Merge pull request #204 from theta42/release-v2.6.0

release(v2.6.0): dedupe promotion groups, resources API perf, nmap RTTVAR fix, real gateway count

.dockerignore

@@ -9,13 +9,23 @@
 .claude
 *.md
 # README.md and tos.md are both read at runtime (tos.md is loaded by
-# routes/index.js at boot), so they must stay in the build context.
+# routes/index.js at boot). DEPLOYMENT.md/API.md/directory_spec.md/docs/*.md
+# are read at runtime too, by routes/docs.js -- all must stay in the build
+# context.
 !README.md
 !tos.md
+!CHANGELOG.md
+!DEPLOYMENT.md
+!API.md
+!directory_spec.md
+!docs/**/*.md
+# The screenshots the README (served at /docs/overview) links. `COPY docs /docs`
+# in Dockerfile.openldap needs these present in the build context.
+!docs/images/**
 
-# Tests
-nodejs/tests/
-nodejs/*.test.js
+# Tests (excluded from production builds; test-runner Dockerfile copies them explicitly)
+# nodejs/tests/
+# nodejs/*.test.js
 
 # Host dependency tree — let the image run a clean `npm ci`. Also avoids
 # copying platform-wrong native modules (e.g. bcrypt built for the host OS).

.gitguardian.yml

@@ -0,0 +1,12 @@
+# GitGuardian configuration (ggshield / GitGuardian GH checks).
+#
+# The generic-password detector false-positives on LDAP admin bind credentials
+# being READ from runtime config (sso-secrets.js / /config/site.json) — e.g.
+# `const x = conf.ldap && conf.ldap.bindPassword` in the multi-site join flow.
+# That is the correct pattern (never a hardcoded secret); ignore the variable
+# reference, not the actual value.
+version: 2
+ignore:
+  - name: generic-password
+    match: |
+      conf\.ldap\s*&&\s*conf\.ldap\.bindPassword

.github/workflows/build-openldap-image.yml

@@ -0,0 +1,58 @@
+name: Build OpenLDAP Base Image
+
+# Publishes ghcr.io/theta42/openldap-nestgroup, the prebuilt slapd-with-
+# nestgroup image Dockerfile.openldap's `ldapbuild` stage pulls FROM instead
+# of compiling from source on every build (see Dockerfile.openldap-builder
+# for why, and the ~5 minute + git.openldap.org-dependent cost it replaces).
+#
+# Runs only when the builder Dockerfile changes -- bumping OPENLDAP_COMMIT in
+# it is the only reason this image should ever need rebuilding -- or on
+# manual dispatch.
+on:
+  push:
+    branches: [master]
+    paths:
+      - 'Dockerfile.openldap-builder'
+      - '.github/workflows/build-openldap-image.yml'
+  workflow_dispatch: {}
+
+permissions:
+  contents: read
+  packages: write
+
+jobs:
+  build-and-push:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      # Single source of truth for the tag: the ARG default in the Dockerfile
+      # itself, not a value duplicated into this workflow.
+      - name: Resolve pinned OpenLDAP commit
+        id: commit
+        run: |
+          commit=$(grep -oP '^ARG OPENLDAP_COMMIT=\K[0-9a-f]+' Dockerfile.openldap-builder)
+          if [ -z "$commit" ]; then
+            echo "::error::Could not resolve OPENLDAP_COMMIT from Dockerfile.openldap-builder"
+            exit 1
+          fi
+          echo "commit=$commit" >> "$GITHUB_OUTPUT"
+
+      - name: Log in to GitHub Container Registry
+        uses: docker/login-action@v2
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Build and push
+        uses: docker/build-push-action@v4
+        with:
+          context: .
+          file: ./Dockerfile.openldap-builder
+          build-args: |
+            OPENLDAP_COMMIT=${{ steps.commit.outputs.commit }}
+          push: true
+          tags: |
+            ghcr.io/theta42/openldap-nestgroup:${{ steps.commit.outputs.commit }}
+            ghcr.io/theta42/openldap-nestgroup:latest

.github/workflows/pr-tests.yml

@@ -0,0 +1,158 @@
+name: Pull Request Tests
+
+# Run tests on pull requests to master and when pushing to PRs
+on:
+  pull_request:
+    branches:
+      - master
+  push:
+    branches-ignore:
+      - master
+
+jobs:
+  test:
+    name: Run Tests
+    runs-on: ubuntu-latest
+
+    strategy:
+      matrix:
+        node-version: [18.x, 20.x, 22.x]
+
+    # A dedicated, GHA-managed Redis -- NOT the bundled image's own Redis,
+    # which only binds to loopback *inside* its container (redis-server's
+    # default with no --bind override), so Docker's -p port-forward can
+    # never actually reach it from the runner. This service container binds
+    # correctly and is reachable at localhost:6379, matching model-redis's
+    # createClient({}) default when conf.redis has no explicit host/port.
+    services:
+      redis:
+        image: redis:7-alpine
+        ports:
+          - 6379:6379
+        options: >-
+          --health-cmd "redis-cli ping"
+          --health-interval 5s
+          --health-timeout 3s
+          --health-retries 5
+
+    steps:
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      # The test suite (require('../app')) also needs a real LDAP directory
+      # seeded with the schema/groups the app expects -- the bundled image
+      # already does exactly that (docker-entrypoint.sh), so build and run
+      # it here rather than reimplementing LDAP setup as a separate
+      # CI-only script. Its own bundled Redis is unused (see services above).
+      - name: Build LDAP test image
+        run: docker build -f Dockerfile.openldap -t sso-test:latest .
+
+      - name: Start LDAP test container
+        run: |
+          mkdir -p /tmp/sso-test-config
+          cp secrets.js.example /tmp/sso-test-config/sso-secrets.js
+          docker run -d --name sso-test \
+            -p 389:389 -p 3001:3001 \
+            -v /tmp/sso-test-config:/config:ro \
+            sso-test:latest
+          for i in $(seq 1 30); do
+            status=$(docker inspect --format='{{.State.Health.Status}}' sso-test 2>/dev/null || echo starting)
+            [ "$status" = "healthy" ] && break
+            sleep 2
+          done
+          docker inspect --format='{{.State.Health.Status}}' sso-test
+
+      # tests/setup.js logs in as uid 'test'; several suites (group/otp/
+      # impersonate/cache) assume a second, non-admin user 'wmantly' already
+      # exists (documented in those test files: "wmantly is always present
+      # in the test LDAP"). Seed both here so CI matches that assumption.
+      - name: Seed test fixtures
+        run: |
+          HASH_TEST=$(timeout 20 docker exec sso-test node -e "console.log(require('/app/models/user_ldap.js').hashPasswordSSHA512('MyTestPassword!2'))" | tail -1)
+          HASH_WMANTLY=$(timeout 20 docker exec sso-test node -e "console.log(require('/app/models/user_ldap.js').hashPasswordSSHA512('WmantlyPass!2'))" | tail -1)
+          cat > /tmp/seed.ldif <<EOF
+          dn: cn=test,ou=people,dc=example,dc=com
+          objectClass: inetOrgPerson
+          objectClass: posixAccount
+          objectClass: theta42Person
+          cn: test
+          sn: Test
+          mail: test@example.com
+          uid: test
+          uidNumber: 10000
+          gidNumber: 10000
+          homeDirectory: /home/test
+          userPassword: ${HASH_TEST}
+          dateOfBirth: 2000-01-01
+
+          dn: cn=app_sso_admin,ou=groups,dc=example,dc=com
+          changetype: modify
+          add: member
+          member: cn=test,ou=people,dc=example,dc=com
+
+          dn: cn=app_sso_oauth_admin,ou=groups,dc=example,dc=com
+          changetype: modify
+          add: member
+          member: cn=test,ou=people,dc=example,dc=com
+
+          dn: cn=app_sso_invite,ou=groups,dc=example,dc=com
+          changetype: modify
+          add: member
+          member: cn=test,ou=people,dc=example,dc=com
+
+          dn: cn=wmantly,ou=people,dc=example,dc=com
+          objectClass: inetOrgPerson
+          objectClass: posixAccount
+          objectClass: theta42Person
+          cn: wmantly
+          sn: Mantly
+          mail: wmantly@example.com
+          uid: wmantly
+          uidNumber: 10001
+          gidNumber: 10001
+          homeDirectory: /home/wmantly
+          userPassword: ${HASH_WMANTLY}
+          dateOfBirth: 2000-01-01
+          EOF
+          docker cp /tmp/seed.ldif sso-test:/tmp/seed.ldif
+          docker exec sso-test ldapmodify -x -D "cn=admin,dc=example,dc=com" -w 'your-ldap-password' -a -f /tmp/seed.ldif
+
+      - name: Setup Node.js ${{ matrix.node-version }}
+        uses: actions/setup-node@v4
+        with:
+          node-version: ${{ matrix.node-version }}
+          cache: 'npm'
+          cache-dependency-path: nodejs/package-lock.json
+
+      - name: Install dependencies
+        working-directory: ./nodejs
+        run: npm ci
+
+      - name: Run tests
+        working-directory: ./nodejs
+        env:
+          NODE_ENV: test
+          # conf/base.js's ldap.* defaults already match secrets.js.example's
+          # directory layout (dc=example,dc=com) -- only the admin password
+          # (normally supplied via a gitignored secrets.js) needs setting.
+          app_ldap__bindPassword: your-ldap-password
+          # routes/oauth.js now refuses to start without a real jwtSecret.
+          # This is a non-secret test value; the container under test uses
+          # secrets.js.example's jwtSecret independently.
+          app_oauth__jwtSecret: ci-test-jwt-secret-do-not-use-in-production
+        run: npm test
+
+  test-summary:
+    name: Test Summary
+    runs-on: ubuntu-latest
+    needs: test
+    if: always()
+
+    steps:
+      - name: Check test results
+        run: |
+          if [ "${{ needs.test.result }}" != "success" ]; then
+            echo "Tests failed. PR cannot be merged."
+            exit 1
+          fi
+          echo "All tests passed successfully!"

.gitignore

@@ -86,6 +86,17 @@ ops/cookbooks/vendor
 secrets.json
 secrets.js
 
+# Per-deployment secret files (real LDAP/SMTP/jwtSecret + generated OAuth
+# creds). theta-env bind-mounts ./config and generates/fills these at setup;
+# they must never be committed. The empty *.example templates ARE tracked.
+config/*-secrets.js
+
+# Default sqlite ORM storage (nodejs/models/index.js falls back to this path
+# when no external DB is configured via conf.orm) -- live runtime data, not a
+# fixture. Was committed by mistake across many prior releases. NB: this is
+# nodejs/config/, distinct from the root ./config/ secrets dir above.
+nodejs/config/*.sqlite
+
 # Jekyll build artifact (GitHub Pages builds remotely; ignore locally)
 docs/_site
 

API.md

@@ -1,5 +1,10 @@
 # SSO Manager API Documentation
 
+> Looking for a plainer explanation of what API tokens are and when you'd
+> want one, instead of a full endpoint reference? See
+> [API Tokens](/docs/api-tokens) (in-app) or
+> [concepts-api-tokens.md](docs/concepts-api-tokens.md) (repo).
+
 ## Overview
 
 API documentation for the SSO Manager Node application. Provides endpoints for authentication, user management, group management, token management, notifications, and OAuth 2.0 / OpenID Connect.
@@ -698,6 +703,10 @@ The authenticated user is automatically set as the group owner.
 { "results": true, "message": "Added user uid to group group." }
 ```
 
+Returns `409` if the user is already a member — common in practice, since
+`groupOfNames` requires at least one member and so seeds whoever created the
+group into it.
+
 ---
 
 ### Remove User from Group
@@ -711,6 +720,66 @@ The authenticated user is automatically set as the group owner.
 
 ---
 
+### Nest a Group Inside Another
+
+**`PUT /api/group/:group/nested/:child`** — `app_sso_admin` or group owner
+
+Makes `:child` a member of `:group`, so everyone in `:child` is a member of
+`:group` at any depth.
+
+**Response:**
+```json
+{ "results": { "cn": "group", "member": ["..."] }, "message": "Nested child inside group." }
+```
+
+**Errors:**
+
+| Status | When |
+|--------|------|
+| `400` | `:group` and `:child` are the same group |
+| `409` | already nested, or the nesting would create a loop (`:child` already contains `:group`, directly or transitively) |
+
+---
+
+### Un-nest a Group
+
+**`DELETE /api/group/:group/nested/:child`** — `app_sso_admin` or group owner
+
+**Response:**
+```json
+{ "results": { "cn": "group", "member": ["..."] }, "message": "Removed child from group." }
+```
+
+**Errors:**
+
+| Status | When |
+|--------|------|
+| `409` | `:child` is the only member — `groupOfNames` requires at least one |
+
+---
+
+### Effective Membership
+
+**`GET /api/group/:group/effective`** — Any authenticated user
+
+Who a group actually grants. `direct` is users listed on the group itself
+(never groups); `nestedGroups` is what is nested into it; `effective` is every
+user reachable through the whole chain.
+
+**Response:**
+```json
+{
+  "results": {
+    "cn": "app_gitea_access",
+    "direct": ["cn=alice,ou=people,dc=example,dc=com"],
+    "nestedGroups": [{ "cn": "developers", "dn": "cn=developers,ou=groups,dc=example,dc=com" }],
+    "effective": ["cn=alice,ou=people,dc=example,dc=com", "cn=bob,ou=people,dc=example,dc=com"]
+  }
+}
+```
+
+---
+
 ### Delete Group
 
 **`DELETE /api/group/:group`** — `app_sso_admin` or group owner
@@ -1130,6 +1199,215 @@ Configurable per-client via `token_lifetime`. Global defaults (in seconds):
 
 ---
 
+## Plugin Endpoints
+
+Base path: `/api/plugins`
+
+All endpoints require authentication and `app_sso_admin`, `app_sso_directory_admin`, or `app_super_admin` membership. Secret field values are always returned masked (`********`); they are stored in OpenBao at `secret/plugins/<instance-id>/conf`, never in the database row. See [Plugins](docs/plugins.html).
+
+### List Plugin Types
+
+**`GET /api/plugins/types`**
+
+Returns the installed plugin types and their `configSchema` (used to build the create-instance form).
+
+**Response:**
+```json
+{
+  "results": [
+    {
+      "type": "proxmox",
+      "category": "discovery",
+      "name": "Proxmox VE",
+      "description": "Discover VMs, containers, and hypervisor nodes from a Proxmox VE API endpoint.",
+      "configSchema": [
+        { "key": "url", "label": "API URL", "type": "url", "required": true },
+        { "key": "tokenId", "label": "Token ID", "type": "text", "required": true },
+        { "key": "tokenSecret", "label": "Token Secret", "type": "password", "required": true, "secret": true }
+      ]
+    }
+  ]
+}
+```
+
+---
+
+### List Plugin Instances
+
+**`GET /api/plugins/`**
+
+**Response:** `{ "results": [ { "id", "pluginType", "category", "name", "slug", "enabled", "cron", "config", "secrets": {…masked…}, "lastRunAt", "lastStatus", "lastError" } ] }`
+
+---
+
+### Get One Instance
+
+**`GET /api/plugins/:id`** — same shape as a list entry.
+
+---
+
+### Create Instance
+
+**`POST /api/plugins/`**
+
+`config` is a flat object of **all** field values (secret and non-secret); the server splits it — non-secret fields go to the DB row, secret fields to OpenBao. Creating an enabled instance schedules it and kicks one immediate run. `slug` is the discovery source name (lowercase letters/digits/_/-, max 64, unique).
+
+**Request:**
+```json
+{
+  "pluginType": "proxmox",
+  "name": "Proxmox — Home Lab",
+  "slug": "proxmox-homelab",
+  "cron": "0 * * * *",
+  "config": { "url": "https://pve:8006", "tokenId": "u@pam!t", "tokenSecret": "secret-value" }
+}
+```
+
+Errors: `400` if the plugin type is unknown, the slug is malformed/duplicated, or a required field is missing; `400` with an OpenBao hint if writing the secret fails (re-run `./setup.sh` with theta-suite ≥ v1.30.1).
+
+---
+
+### Update Instance
+
+**`PUT /api/plugins/:id`** — update `name`, `cron`, `enabled`, and non-secret `config`. Secret fields are changed via `PUT /:id/secrets`. Re-schedules if `cron` or `enabled` changed.
+
+---
+
+### Update Secrets
+
+**`PUT /api/plugins/:id/secrets`** — body is a flat object of secret field values. Blank/`********` values are ignored (kept as-is).
+
+---
+
+### Test Instance
+
+**`POST /api/plugins/:id/test`** — runs the plugin's `validate`. Returns `{ "ok": true }` or `400 { "ok": false, "error": "..." }`.
+
+---
+
+### Load / Unload / Run Now
+
+- **`POST /api/plugins/:id/load`** — enable + schedule + run now.
+- **`POST /api/plugins/:id/unload`** — unschedule + disable.
+- **`POST /api/plugins/:id/run`** — enqueue one immediate run (regardless of enabled).
+
+---
+
+### Last Run Status
+
+**`GET /api/plugins/:id/runs`** → `{ "results": { "lastRunAt", "lastStatus", "lastError" } }` (`lastStatus` is `ok` | `error` | `running`).
+
+---
+
+### Delete Instance
+
+**`DELETE /api/plugins/:id`** — unschedules, removes the OpenBao secret namespace, and deletes the row.
+
+## Configuration Endpoints
+
+Base path: `/api/conf`
+
+All endpoints require authentication and `app_sso_admin` membership. Runtime configuration (SMTP, discovery, OAuth) is stored in OpenBao at `secret/sso-manager/conf` and overlaid onto the live app config; changes take effect immediately and persist across restarts. Secret fields (`smtp.pass`, `oauth.jwtSecret`) are **always returned masked** (`********`); submit a blank or `********` value to keep the current stored secret, or a new non-blank value to replace it.
+
+### Get Configuration
+
+**`GET /api/conf`** — returns the editable config groups (`smtp`, `discovery`, `oauth`) with secret fields masked to `********`.
+
+**Response:**
+```json
+{
+  "smtp": { "host": "smtp.example.com", "port": 587, "secure": false, "user": "noreply@example.com", "pass": "********", "from": "SSO Manager <noreply@example.com>" },
+  "discovery": { },
+  "oauth": { "issuer": "https://sso.example.com", "jwtSecret": "********", "token_lifetime": { "access_token": 3600, "refresh_token": 2592000 } }
+}
+```
+
+### Save Configuration
+
+**`POST /api/conf`** — deep-merges the submitted groups into `secret/sso-manager/conf` (per-key shallow merge of nested objects) and re-applies them to the live config. A blank or `********` value for `smtp.pass` or `oauth.jwtSecret` preserves the stored secret.
+
+**Request:**
+```json
+{
+  "smtp": { "host": "smtp.example.com", "port": 587, "secure": false, "user": "noreply@example.com", "pass": "********", "from": "SSO Manager <noreply@example.com>" },
+  "oauth": { "issuer": "https://sso.example.com", "token_lifetime": { "access_token": 3600, "refresh_token": 2592000 } }
+}
+```
+
+**Response:** `{ "success": true }`
+
+---
+
+## Subtype Driver Operations Endpoints
+
+Base path: `/api/directory-admin/resources`
+
+All endpoints require authentication and `app_sso_admin`, `app_sso_directory_admin`, or `admin` permission.
+
+### Get Subtype Driver Metrics
+
+**`GET /api/directory-admin/resources/:id/driver-metrics`**
+
+Resolves the operational driver for the resource via the 4-tier engine (`theta-agent`, specialized subtype driver, parent hypervisor provider, or unmanaged fallback) and returns real-time telemetry.
+
+**Response:**
+```json
+{
+  "status": "ok",
+  "resourceId": "res-id",
+  "metrics": {
+    "status": "online",
+    "driver": "database",
+    "subType": "redis",
+    "redis": { "connectedClients": 4, "usedMemoryBytes": 12582912, "opsPerSec": 42 }
+  }
+}
+```
+
+---
+
+### Execute Subtype Driver Action
+
+**`POST /api/directory-admin/resources/:id/driver-action`**
+
+Executes a protocol action on the target resource (e.g. systemd restart, Proxmox power control, Redis flush, K8s scale).
+
+**Request:**
+```json
+{
+  "action": "restart",
+  "params": { "serviceName": "emby-server" }
+}
+```
+
+**Response:**
+```json
+{
+  "status": "ok",
+  "resourceId": "res-id",
+  "result": { "status": "ok", "driver": "docker_socket", "action": "restart" }
+}
+```
+
+---
+
+### Get Subtype Driver Logs
+
+**`GET /api/directory-admin/resources/:id/driver-logs?lines=100`**
+
+Retrieves recent operational logs for the resource via the resolved driver (`journalctl`, `docker logs`, Proxmox task logs, K8s pod logs).
+
+**Response:**
+```json
+{
+  "status": "ok",
+  "resourceId": "res-id",
+  "logs": "[docker logs --tail 100 emby-server]\nContainer initialized..."
+}
+```
+
+---
+
 ## Error Responses
 
 All endpoints return errors in this format:

DEPLOYMENT.md

@@ -120,7 +120,7 @@ bare-metal / advanced standalone use; most deployments should use the file.
 - Health check: `http://localhost:3001/health` → `{"status":"ok"}`
 - OIDC discovery: `http://localhost:3001/.well-known/openid-configuration`
 - LDAP (internal, app↔slapd): `ldap://localhost:389` (not mapped to the host)
-- LDAPS (for legacy apps / direct binds): `ldaps://<host>:636` (TLS)
+- LDAPS (direct binds: Linux hosts, LDAP-native apps): `ldaps://<host>:636` (TLS)
 
 ### API tokens (personal access tokens)
 
@@ -176,6 +176,8 @@ docker compose exec sso-manager ldapsearch -x -H ldap://localhost:389 \
 | `PORT` | `3001` | host port mapped to the UI |
 | `LDAPS_PORT` | `636` | host port mapped to LDAPS |
 | `LDAP_PORT` | `389` | uncomment the host mapping in compose to expose plain LDAP (not recommended) |
+| `LDAP_SERVER_ID` | empty | Unique integer ID (e.g. 1, 2) required to enable Multi-Master replication |
+| `LDAP_REPLICATION_HOSTS` | empty | Space-separated list of other sites' LDAP URLs for replication (e.g. `ldaps://site2:636`) |
 
 Any `app_*` var may also be set directly to override any config value (see the
 table at the top).
@@ -188,6 +190,12 @@ valid 10 years, SAN includes the CN + `localhost` + `127.0.0.1`) and listens on
 `ldap-certs` volume so it persists across container recreation — clients don't need
 to re-trust on every rebuild.
 
+The `/integrations` page derives its LDAPS URL from the OAuth issuer by default.
+To advertise a separate, internal-only hostname (e.g. `ldap.internal.example.com`
+or `sso-manager` for Docker-internal clients), set `conf.ldap.ldapsHost` in your
+secrets file or pass `app_ldap__ldapsHost=...`. See `docs/ldap.md` for
+recommended network layouts and how to match the cert SAN to the hostname.
+
 - **Trusting the self-signed cert** (clients): copy `/etc/openldap/certs/ldap.crt`
   out of the container and add it to the client's trusted CA store, or set
   `TLS_REQCERT never` for quick-and-dirty LAN use. Fetch it with:
@@ -328,14 +336,32 @@ OAuth client). Note: re-running bootstrap resets the bootstrap-admin and
 service-account passwords to the values in `./config/sso-secrets.js`; non-theta
 OAuth clients live in SSO Redis and are preserved by the volume.
 
+> **Note — the bundled slapd is built from source.** The all-in-one image
+> compiles OpenLDAP from a pinned upstream commit to get the `nestgroup`
+> overlay (nested groups; see `docs/directory.md`), because no 2.6.x release
+> ships it. One consequence: master uses **LMDB 1.0.0**, whose on-disk format is
+> mutually unreadable with the 0.9.x in OpenLDAP 2.6.x
+> (`MDB_INVALID: File is not an LMDB file`). Moving a directory between a 2.6.x
+> image and this one is a `slapcat` → `slapadd` reload, not a restart — the same
+> shape as "Restore — LDAP only" above. Verify after a rebuild:
+> `docker compose logs sso-manager | grep nestgroup` should report the overlay
+> as available.
+
 ---
 
 ## Method 2: Bare metal (Debian/Ubuntu)
 
-`install.sh` is an idempotent installer: it installs Node.js 20.x, installs and
-configures OpenLDAP (modules + overlays + custom schema + directory tree +
-required groups), deploys the app to `/opt/sso-manager`, and creates a systemd
-unit. Configuration is written to `/opt/sso-manager/conf/secrets.js` (file-based).
+`install.sh` is an idempotent installer: it installs Node.js 22.x and Redis,
+force-syncs the repo to `/opt/theta42/sso-manager`, and symlinks the systemd
+config from the repo. Re-run it to update — it prints the version you're
+updating from and to (or "Already up to date" if there's nothing new).
+
+On the **first run only** it also installs and configures OpenLDAP (modules +
+overlays + custom schema + directory tree + required groups — see
+`ops/ldap-setup.sh`) and seeds `/etc/sso-manager/secrets.js` with a generated
+LDAP admin password and JWT secret (SMTP is left as a placeholder). Once that
+file exists it's never touched again, and LDAP is never re-bootstrapped —
+edit the file and restart the service to change anything.
 
 ### Prerequisites
 
@@ -346,47 +372,50 @@ unit. Configuration is written to `/opt/sso-manager/conf/secrets.js` (file-based
 ### Install
 
 ```bash
-sudo ./install.sh \
-  -p 'your-ldap-password' \
-  -b 'dc=yourdomain,dc=com' \
-  -n 'Your Org' \
-  -o 3001
+wget -O - https://raw.githubusercontent.com/theta42/sso-manager-node/master/install.sh | sudo bash
 ```
 
-| Flag | Env var | Description |
-|------|---------|-------------|
-| `-p, --admin-pass` | `LDAP_ADMIN_PASS` | LDAP admin password (required) |
-| `-b, --base-dn` | `LDAP_BASE_DN` | Base DN (default `dc=example,dc=com`) |
-| `-n, --org-name` | `ORG_NAME` | Org name (default `SSO Manager`) |
-| `-o, --port` | `PORT` | HTTP port (default `3001`) |
-| `-j, --jwt-secret` | `JWT_SECRET` | JWT secret (default auto-generated) |
-| `-s, --smtp-config` | `SMTP_*` | SMTP as `host:port:user:pass` |
-| `--skip-ldap` | `SKIP_LDAP` | Skip LDAP setup (use existing) |
-| `--skip-app` | `SKIP_APP` | LDAP setup only |
-| `--dry-run` | `DRY_RUN` | Show actions without making changes |
+or, if you already have the repo checked out:
+
+```bash
+sudo ./install.sh
+```
+
+| Env var | Description |
+|---------|-------------|
+| `LDAP_BASE_DN` | Base DN (default `dc=example,dc=com`) — first run only |
+| `LDAP_ADMIN_PASS` | LDAP admin password (default auto-generated) — first run only |
+| `JWT_SECRET` | JWT secret (default auto-generated) — first run only |
+| `ORG_NAME` | Org name (default `SSO Manager`) — first run only |
+| `PORT` | HTTP port (default `3001`) — first run only |
+| `SKIP_LDAP` | `true` to skip OpenLDAP bootstrap entirely (point at an existing server yourself) |
+| `REPO_URL`, `REPO_DIR`, `BRANCH`, `SECRETS_FILE` | Override the defaults |
 
 ### Post-install
 
 ```bash
-sudo systemctl enable --now sso-manager
+sudo systemctl status sso-manager
 journalctl -fu sso-manager
 curl http://localhost:3001/health    # -> {"status":"ok"}
 ```
 
 ### What `install.sh` does
 
-1. Installs Node.js 20.x (NodeSource).
-2. Installs OpenLDAP (`slapd`) with: `pw-sha2`, `ppolicy`, `memberof`, `refint`
-   modules + overlays; the custom `theta42Person` schema (`dateOfBirth`); indexes;
-   `ou=people`/`ou=groups`/`ou=policies`; a default `pwdPolicy`; and the SSO groups.
-3. Installs the app to `/opt/sso-manager` and runs `npm ci --omit=dev`.
-4. Generates `conf/secrets.js` (LDAP/SMTP/JWT) and `conf/base.js` (generic defaults).
-5. Installs `sso-manager.service` (systemd), enabled on boot.
+1. Installs Node.js 22.x (NodeSource) and Redis.
+2. Clones/updates the repo at `/opt/theta42/sso-manager`.
+3. **First run only:** installs OpenLDAP (`slapd`) with `pw-sha2`, `ppolicy`,
+   `memberof`, `refint` modules + overlays; the custom `theta42Person` schema
+   (`dateOfBirth`); indexes; `ou=people`/`ou=groups`/`ou=policies`; a default
+   `pwdPolicy`; and the SSO groups — then seeds `/etc/sso-manager/secrets.js`.
+4. Symlinks `ops/systemd/sso-manager.service` into `/etc/systemd/system` and
+   runs `npm ci --omit=dev`.
+5. Enables and (re)starts the service.
 
-> For an existing LDAP server, run `sudo ./install.sh --skip-ldap …` and point the
-> app at it. For LDAP-only setup on a host that already runs the app elsewhere, use
-> `--skip-app`. To (re)configure overlays on an already-installed slapd, prefer
-> `ops/ldap-setup.sh` (idempotent, auto-detects the user database).
+> For an existing LDAP server, run with `SKIP_LDAP=true` and write
+> `/etc/sso-manager/secrets.js` yourself (see `secrets.js.example`) before
+> starting the service. To (re)configure overlays on an already-installed
+> slapd, use `ops/ldap-setup.sh` directly (idempotent, auto-detects the user
+> database).
 
 ---
 
@@ -465,8 +494,8 @@ netstat -tlnp | grep 389
 2. **Use LDAPS / StartTLS** for any LDAP connection that crosses the network. The
    bundled slapd listens on `ldaps:///` (636, TLS) and `ldap:///` (389, plain +
    StartTLS); port 389 is not mapped to the host by default so LAN clients can't
-   bind in cleartext. Direct-LDAP apps (legacy services, `theta42/proxy`) should
-   use `ldaps://…:636` or StartTLS.
+   bind in cleartext. Direct-LDAP consumers (Linux hosts, LDAP-native apps,
+   `theta42/proxy`) should use `ldaps://…:636` or StartTLS.
 3. **Persist `JWT_SECRET`** — if the Docker image auto-generates one and you don't
    set `JWT_SECRET`, issued tokens invalidate on container recreation.
 4. **Don't expose the UI's HTTP port to the internet** — terminate TLS at a front

Dockerfile.openldap

@@ -22,6 +22,12 @@
 # GIT_COMMIT=$(git -C sso-manager-node rev-parse --short HEAD), computed on
 # the host where the submodule resolves correctly.
 ARG GIT_COMMIT=""
+# Pinned OpenLDAP commit this build expects -- must match the tag of the
+# published builder image below. Bumping it is a two-step change: rebuild +
+# push ghcr.io/theta42/openldap-nestgroup:<new commit> from
+# Dockerfile.openldap-builder (see that file), then update this default (or
+# pass --build-arg OPENLDAP_COMMIT=<new commit> here).
+ARG OPENLDAP_COMMIT=350e9eb38b2270c2bad97c61ee02e85fb8f3196d
 FROM node:20-alpine AS gitinfo
 ARG GIT_COMMIT
 WORKDIR /repo
@@ -33,37 +39,69 @@ RUN if [ -n "$GIT_COMMIT" ]; then \
         && git rev-parse --short HEAD > /commit.txt; } 2>/dev/null || echo unknown > /commit.txt; \
     fi
 
+# ── OpenLDAP, prebuilt ────────────────────────────────────────────────────────
+# We run slapd from OpenLDAP master rather than Alpine's packaged release, for
+# exactly one feature: the `nestgroup` overlay (ITS#10161, Howard Chu,
+# 2024-03-21), which evaluates nested groups server-side. Nothing in any 2.6.x
+# release can do this -- verified: 2.6.13 ships 26 overlay modules and
+# nestgroup is not among them -- and the alternative is resolving nesting
+# separately in every consumer (this app, SSSD on each host, jump-host, proxy),
+# where any consumer that forgets silently under-grants access.
+#
+# Consequence to know about: master ships LMDB 1.0.0, whose on-disk format the
+# 0.9.x used by 2.6.x cannot read, and vice versa
+# ("MDB_INVALID: File is not an LMDB file"). Moving an existing directory onto
+# this image is a slapcat/slapadd migration, not a restart. See DEPLOYMENT.md.
+#
+# The from-source compile (~5 min, and a dependency on git.openldap.org being
+# reachable) used to happen right here, on every build of this Dockerfile --
+# including 3x per CI run's test matrix. It's now built once, tagged by the
+# pinned commit above, in Dockerfile.openldap-builder -- see that file for the
+# actual compile steps and the TODO on dropping from-source entirely once
+# nestgroup ships in a release.
+FROM ghcr.io/theta42/openldap-nestgroup:${OPENLDAP_COMMIT} AS ldapbuild
+
 FROM node:20-alpine
 
-# Install OpenLDAP and required packages.
-# Alpine splits OpenLDAP into many small subpackages; there is no catch-all
-# "openldap-overlays" package. We install exactly the backends/overlays/modules
-# the app depends on:
-#   openldap-back-mdb     : the mdb backend (slapd.conf uses `database mdb`)
-#   openldap-overlay-ppolicy   : ppolicy module + overlay (account locking)
-#   openldap-overlay-memberof  : reverse group membership
-#   openldap-overlay-refint    : referential integrity on group members
-#   openldap-passwd-sha2       : pw-sha2 module ({SSHA512} user password hashing)
-# Note: Alpine does NOT ship a ppolicy.schema file — on OpenLDAP 2.6 the ppolicy
-# schema is built into ppolicy.so and registered when the module loads, so
-# docker-entrypoint.sh loads it via `moduleload ppolicy` (no schema include).
-#   openssl : used by docker-entrypoint.sh to generate a JWT secret
+# Runtime libraries the from-source slapd links against, plus the app's own
+# deps. No openldap* packages here: everything LDAP comes from /opt/openldap.
+# libltdl (module loading -- slapd is useless without it, since every overlay
+# is a loadable module) and libuuid are pulled in by the source build but are
+# NOT dependencies of anything else here, so they must be named explicitly;
+# omitting them fails at runtime with "Error relocating ... lt_dlopenext:
+# symbol not found", not at build time.
 RUN apk add --no-cache \
-    openldap \
-    openldap-clients \
-    openldap-back-mdb \
-    openldap-overlay-ppolicy \
-    openldap-overlay-memberof \
-    openldap-overlay-refint \
-    openldap-passwd-sha2 \
+    openssl \
+    libsasl \
+    libltdl \
+    libuuid \
     dumb-init \
     bash \
-    openssl \
     redis \
+    nmap \
     && rm -rf /var/cache/apk/*
 
-# The openldap package already creates the `ldap` user/group, which slapd runs
-# as (see -u ldap -g ldap in docker-entrypoint.sh). Nothing to add here.
+COPY --from=ldapbuild /opt/openldap /opt/openldap
+# Which upstream commit this slapd was built from — so a running container can
+# answer "what am I actually running" without rebuilding.
+COPY --from=ldapbuild /opt-openldap-commit.txt /opt/openldap/COMMIT
+
+# The Alpine openldap package used to create these; nothing does now, and
+# docker-entrypoint.sh runs slapd as -u ldap -g ldap.
+RUN addgroup -S ldap 2>/dev/null || true \
+    && adduser -S -D -H -G ldap ldap 2>/dev/null || true
+
+# docker-entrypoint.sh invokes slapd/slappasswd/ldapadd/ldapsearch by bare name
+# and probes a list of candidate module directories, so putting the from-source
+# tree first on PATH is all that is needed to redirect it. Schemas are symlinked
+# into the conventional location because the entrypoint's slapd.conf includes
+# /etc/openldap/schema/*.schema, and the app's own schemas (theta42, sudo,
+# openssh-lpk) are copied there too.
+ENV PATH="/opt/openldap/bin:/opt/openldap/sbin:/opt/openldap/libexec:${PATH}"
+RUN mkdir -p /etc/openldap/schema \
+    && for f in /opt/openldap/etc/openldap/schema/*.schema; do \
+         ln -sf "$f" "/etc/openldap/schema/$(basename "$f")"; \
+       done
 
 WORKDIR /app
 
@@ -81,6 +119,7 @@ COPY nodejs/app.js ./
 COPY nodejs/bin ./bin
 COPY nodejs/conf ./conf
 COPY nodejs/controller ./controller
+COPY nodejs/drivers ./drivers
 COPY nodejs/middleware ./middleware
 COPY nodejs/models ./models
 COPY nodejs/routes ./routes
@@ -88,6 +127,7 @@ COPY nodejs/services ./services
 COPY nodejs/utils ./utils
 COPY nodejs/views ./views
 COPY nodejs/public ./public
+COPY nodejs/plugins ./plugins
 
 # routes/index.js reads path.join(__dirname, '../../tos.md') at boot. With the
 # app flattened into /app, __dirname is /app/routes and ../../ resolves to /,
@@ -95,6 +135,18 @@ COPY nodejs/public ./public
 # level above the nodejs/ app dir). Without this the app crashes on startup.
 COPY tos.md /tos.md
 
+# Documentation, served in-app at /docs (routes/docs.js) so it's readable
+# without internet access. Same flattened-path convention as tos.md above.
+COPY README.md /README.md
+COPY CHANGELOG.md /CHANGELOG.md
+COPY API.md /API.md
+COPY directory_spec.md /directory_spec.md
+# The docs/*.md tree (plus the images the docs link) is read at runtime too, so
+# the whole docs/ dir must land at /docs. Without this every in-app /docs/<slug>
+# page other than the root-level README/CHANGELOG/API/directory_spec 500s on the
+# fs.readFileSync in routes/docs.js (files missing from the image).
+COPY docs /docs
+
 # Baked commit hash from the gitinfo stage (see build_info.js).
 COPY --from=gitinfo /commit.txt ./.build_commit
 
@@ -120,7 +172,8 @@ COPY ops/schema/openssh-lpk.schema /etc/openldap/schema/openssh-lpk.schema
 # 3001: SSO Manager web interface (HTTP — terminate TLS at the front proxy)
 # 389:  LDAP  (plain + StartTLS) — used internally by the app; map to host only
 #       if you want LAN clients to bind without TLS (not recommended).
-# 636:  LDAPS — for legacy apps / direct LDAP binds over the network (TLS)
+# 636:  LDAPS — direct LDAP binds over the network (TLS): Linux host auth
+#       (PAM/SSSD, sudo, SSH keys) and LDAP-native apps
 EXPOSE 3001 389 636
 
 # Health check

Dockerfile.openldap-builder

@@ -0,0 +1,75 @@
+# OpenLDAP-with-nestgroup builder, published to
+# ghcr.io/theta42/openldap-nestgroup:<OPENLDAP_COMMIT short hash>.
+#
+# Extracted out of Dockerfile.openldap's `ldapbuild` stage so the ~5 minute
+# from-source compile (which also depends on git.openldap.org being up)
+# happens once, here, instead of on every `docker build` of the app image --
+# including every CI run's 3-way test matrix. Dockerfile.openldap's ldapbuild
+# stage becomes `FROM ghcr.io/theta42/openldap-nestgroup:<commit>` and the
+# rest of that file (the COPY --from=ldapbuild lines) is unchanged, since
+# COPY --from also accepts an external image, not just a local stage name.
+#
+# Bumping OPENLDAP_COMMIT is a two-step change: update the ARG below, push
+# (the build-openldap-image workflow rebuilds+republishes the tag on changes
+# to this file), then update the matching FROM line in Dockerfile.openldap.
+#
+# See Dockerfile.openldap's own "OpenLDAP from source" comment for *why*
+# from-source at all (the nestgroup overlay, ITS#10161) and the LMDB format
+# note (master's 1.0.0 vs 2.6.x's 0.9.x).
+FROM node:20-alpine AS build
+
+# groff is not optional despite producing nothing we ship: the build descends
+# into doc/man unconditionally and its Makefile calls soelim, which groff
+# provides. Without it the whole `make` fails at the man-page stage
+# ("soelim: not found") long after slapd itself has compiled fine.
+RUN apk add --no-cache \
+        build-base autoconf automake libtool \
+        openssl-dev cyrus-sasl-dev \
+        git make pkgconf util-linux-dev groff
+
+# Pinned to an exact commit, not a branch tip -- see Dockerfile.openldap for
+# why (this is the directory server the whole lab authenticates against).
+ARG OPENLDAP_COMMIT=350e9eb38b2270c2bad97c61ee02e85fb8f3196d
+
+WORKDIR /src
+RUN git init -q . \
+    && git remote add origin https://git.openldap.org/openldap/openldap.git \
+    && git fetch -q --depth 1 origin "${OPENLDAP_COMMIT}" \
+    && git checkout -q FETCH_HEAD \
+    && git rev-parse HEAD > /opt-openldap-commit.txt
+
+# Overlays are built as loadable modules (=mod) because docker-entrypoint.sh
+# `moduleload`s them individually; nestgroup joins that set.
+RUN ./configure \
+        --prefix=/opt/openldap \
+        --enable-slapd \
+        --enable-modules \
+        --enable-mdb \
+        --enable-memberof=mod \
+        --enable-refint=mod \
+        --enable-ppolicy=mod \
+        --enable-dynlist=mod \
+        --enable-nestgroup=mod \
+        --enable-syncprov=mod \
+        --enable-auditlog=mod \
+        --with-tls=openssl \
+        --with-cyrus-sasl \
+    && make depend \
+    && make -j"$(nproc)" \
+    && make install
+
+# pw-sha2 provides {SSHA512}, which every existing user password is stored as.
+# It lives in contrib and is not covered by the configure flags above, so it is
+# built separately against the just-built tree -- omitting it would make every
+# user password unverifiable.
+RUN cd contrib/slapd-modules/passwd/sha2 \
+    && make prefix=/opt/openldap OPENLDAP_SRC=/src \
+    && cp .libs/pw-sha2.so* /opt/openldap/libexec/openldap/
+
+# Pure artifact holder -- no shell, no package manager, nothing but the
+# compiled tree. Dockerfile.openldap's COPY --from=ldapbuild only ever reads
+# files, never RUNs anything in this stage, so scratch is sufficient and
+# keeps the published image (and every pull of it) as small as possible.
+FROM scratch
+COPY --from=build /opt/openldap /opt/openldap
+COPY --from=build /opt-openldap-commit.txt /opt-openldap-commit.txt

Dockerfile.test-runner

@@ -0,0 +1,57 @@
+# Test-runner image for SSO Manager.
+#
+# Installs all dependencies (including dev) and bundles the app code plus
+# the seed script. The entrypoint waits for LDAP + Redis, seeds the test
+# user, then runs whatever command is given (default: npm test).
+
+FROM node:20-alpine
+
+# Install OpenLDAP clients (ldapadd, ldapsearch) and bash for the seed script
+RUN apk add --no-cache openldap-clients bash
+
+WORKDIR /app
+
+# Copy and install dependencies (including devDependencies for jest/supertest)
+COPY nodejs/package*.json ./
+RUN npm ci
+
+# Copy the application source
+COPY nodejs/app.js ./
+COPY nodejs/bin ./bin
+COPY nodejs/conf ./conf
+COPY nodejs/controller ./controller
+COPY nodejs/drivers ./drivers
+COPY nodejs/middleware ./middleware
+COPY nodejs/models ./models
+# Without this the discovery/plugin suites cannot even load their subject and
+# fail as "Cannot find module ../plugins/discovery/..." -- plugin code was
+# effectively untested in CI.
+COPY nodejs/plugins ./plugins
+COPY nodejs/routes ./routes
+COPY nodejs/services ./services
+COPY nodejs/utils ./utils
+COPY nodejs/views ./views
+COPY nodejs/public ./public
+COPY nodejs/tests ./tests
+
+# SQLite database directory (config/inventory.sqlite for Resource model's ORM)
+RUN mkdir -p /app/config
+
+# Files expected at the flattened /app path (see Dockerfile.openldap notes)
+COPY tos.md /tos.md
+COPY README.md /README.md
+COPY CHANGELOG.md /CHANGELOG.md
+COPY API.md /API.md
+COPY directory_spec.md /directory_spec.md
+
+# Seed script and utility
+COPY test_seed.js ./test_seed.js
+COPY test/seed-test-user.sh /usr/local/bin/seed-test-user
+RUN chmod +x /usr/local/bin/seed-test-user
+# End-to-end LDAP tunnel test client (docker-compose.e2e.yml)
+COPY test/tunnel_e2e.js ./test/tunnel_e2e.js
+# End-to-end multi-site join test client (docker-compose.multisite-e2e.yml)
+COPY test/multisite_join_e2e.js ./test/multisite_join_e2e.js
+
+# Default command: seed the test user, then run the test suite
+CMD ["sh", "-c", "seed-test-user && npm test"]

README.md

@@ -1,19 +1,15 @@
-# SSO Manager
+# Theta Directory
 
-A self-hosted **OpenID Connect provider** with a bundled **OpenLDAP directory**
-and a web management UI — for home labs and small businesses that want their own
-identity provider instead of a hosted one.
+A production-grade, self-hosted **OpenID Connect provider**, **Resource Directory & IAM Engine**, and bundled **OpenLDAP directory** with a modern web console — designed for home-labs and enterprise infrastructure that demand total sovereignty over their identity, secrets, and resource catalog.
 
-It gives you one place to manage your users and groups, one login (OIDC) that
-your modern apps can use, and one LDAP directory your older or odder apps can
-bind to directly. Everything runs on your own hardware; there is no
-phone-home, no hosted control plane, and no per-user pricing.
+It provides a single source of truth for identity (OIDC + LDAP), host/service directory inventory, access control groups, and secrets management running entirely on your own hardware without third-party cloud lock-in.
 
-> Setting up the whole stack (this SSO + the [theta42/proxy](https://github.com/theta42/proxy)
-> in front of it) with one command? Skip to [theta-env](https://github.com/theta42/theta-env)
-> — its `setup.sh` wires the two together and generates the config for you.
+Theta Directory is deployed as part of [Theta Suite](https://github.com/theta42/theta-suite),
+alongside [Theta Proxy](https://github.com/theta42/proxy) and
+[Theta Gateway](https://github.com/theta42/jump-host) — it isn't installed or
+run on its own. `./setup.sh` wires the whole stack together automatically.
 
-**Documentation:** [https://theta42.github.io/sso-manager-node/](https://theta42.github.io/sso-manager-node/)
+**Documentation:** [https://theta42.github.io/theta-suite/sso/](https://theta42.github.io/theta-suite/sso/)
 
 ## Screenshots
 
@@ -25,6 +21,14 @@ phone-home, no hosted control plane, and no per-user pricing.
 | --- | --- |
 | [![Groups](docs/images/groups.png)](docs/images/groups.png) | [![OAuth clients](docs/images/oauth-clients.png)](docs/images/oauth-clients.png) |
 
+| Sites & Replication |
+| --- |
+| [![Sites](docs/images/sites.png)](docs/images/sites.png) |
+
+| Agent Capabilities & Metrics | Agent Install (Join Key) |
+| --- | --- |
+| [![Agent capabilities & metrics](docs/images/agent-capabilities-metrics.png)](docs/images/agent-capabilities-metrics.png) | [![Agent install with join key](docs/images/agent-install-join-key.png)](docs/images/agent-install-join-key.png) |
+
 ## Features
 
 - **OpenID Connect / OAuth 2.0 provider** — issue your own access, refresh, and
@@ -37,93 +41,39 @@ phone-home, no hosted control plane, and no per-user pricing.
 - **Web management UI** — manage users, groups, and OAuth clients from a
   browser; invite and password-reset flows over email; user self-service for
   profile and API tokens.
-- **LDAPS for legacy apps** — apps that bind LDAP directly (Gitea, Emby, and
-  anything else that speaks LDAP) use LDAPS (636) or StartTLS against the same
-  directory, so you don't maintain a second user database for them.
+- **Direct LDAP binds** — Linux hosts (PAM/SSSD login, LDAP-backed `sudo`
+  rules, SSH public keys via openssh-lpk) and LDAP-native apps (Gitea, Emby,
+  and anything else that speaks LDAP) use LDAPS (636) or StartTLS against the
+  same directory, so you don't maintain a second user database for them.
 - **Personal access tokens** — any user can mint a long-lived bearer token to
   drive the management API from scripts or CI, scoped to their own permissions.
-- **All-in-one Docker image** — app + OpenLDAP + Redis in one container, or run
-  the pieces separately against your own LDAP/Redis via `app_*` env config.
+- **Directory & Inventory Graph** — full host/service/site graph with resource metadata, automatic LDAP group provisioning (`_access` / `_admin`), and Access Request workflows.
+- **Subtype Management & Metrics Drivers Engine** — 4-tier resolution engine binding `subType` metadata (`systemd`, `docker`, `proxmox`, `wireguard`, `postgresql`, `redis`, `k8s`) to operational telemetry, log streaming, and remote lifecycle control.
+- **Explicit Secret Inheritance Mode** — OpenBao KV-v2 integration with strict upward ancestor lineage (`Resource -> Host -> Cluster -> Site`), preserving precise secret scoping across services and containers.
+- **Multi-Site Support (Geo-Location Scaling)** — built-in support for N-Way Multi-Master OpenLDAP replication across physical sites for HA and low latency.
 
-## Why this over the alternatives
+## Secrets
 
-Tools like Keycloak, Authentik, Authelia, or Zitadel are OIDC providers, but
-LDAP is either a paid feature, a federation target you have to run separately,
-or absent. If your stack already has apps that speak LDAP directly (or you just
-want one real directory as the source of truth), you end up running *two*
-identity systems and keeping them in sync.
+Secrets are loaded from **OpenBao** at boot via
+[@simpleworkjs/bao-conf](https://simpleworkjs.github.io/bao-conf/), which
+deep-merges `secret/sso-manager/conf` over the file-loaded config (fail-soft:
+if OpenBao is unreachable, boot continues from `CONF_SECRETS`). The SSO
+authenticates to OpenBao with the scoped `VAULT_TOKEN` (env, policy
+`sso-broker`) — never the root token.
 
-SSO Manager bundles the OpenLDAP directory with the OIDC provider, so OIDC
-apps and LDAP apps read from the same users and groups. The trade-off is
-scope: it is intentionally small and self-hosted, not an enterprise IAM suite
-— no fancy workflow engine, no hosted SaaS. If you want a lightweight,
-self-contained identity provider with a real LDAP backend, that is the niche.
+The SSO also acts as the **vault broker** for the whole stack: it mints
+per-user (`user-<uid>`) and per-admin (`sso-admin`) tokens through the
+`sso-broker` token role and exposes the personal-secrets UI at **Vault → My
+Secrets** (`secret/users/<uid>/*`, server-side token injection + path-scope
+guard) and an admin **Apps** tab to mint scoped tokens for external apps
+(`secret/apps/<name>/*`). The old `utils/conf_manager.js` was replaced by
+`@simpleworkjs/bao-conf`; the admin **Configuration** UI (`/api/conf`) now
+writes `secret/sso-manager/conf` through `bao-conf.set`.
 
-## Quick start
-
-Three ways to run it, in order of how much it sets up for you:
-
-### 1. As part of the unified stack (recommended)
-
-[theta-env](https://github.com/theta42/theta-env) composes this SSO Manager with
-the [theta42/proxy](https://github.com/theta42/proxy) (an OIDC-protected reverse
-proxy) and generates all the config from a single `setup.env` — you enter your
-domain once and it fills in the LDAP DNs, hostnames, OAuth issuer, and random
-secrets consistently:
-
-```bash
-git clone --recursive https://github.com/theta42/theta-env.git
-cd theta-env
-cp setup.env.example setup.env   # set CFG_DOMAIN to your domain
-./setup.sh                       # generates ./config/, builds + bootstraps + starts both
-```
-
-See the [theta-env README](https://github.com/theta42/theta-env) for the full
-first-run flow, DNS/port requirements, and backups.
-
-### 2. Standalone, in Docker
-
-The all-in-one image bundles the app, OpenLDAP, and Redis. Copy the example
-secrets file, fill in your values, and build:
-
-```bash
-git clone https://github.com/theta42/sso-manager-node.git
-cd sso-manager-node
-mkdir -p config && chmod 700 config
-cp secrets.js.example config/sso-secrets.js
-$EDITOR config/sso-secrets.js     # set ldap.bindPassword, oauth.jwtSecret, ...
-docker compose up -d --build
-```
-
-The web UI comes up at `http://localhost:3001`. To kick the tires with no
-config file at all, the entrypoint falls back to safe defaults
-(`dc=example,dc=com`, admin password `admin`, an auto-generated JWT) — fine for
-a local test, not for production.
-
-Your domain is entered once, as the LDAP base DN (`stack.ldapBaseDn`); the other
-LDAP DNs and the OAuth issuer derive from it and must stay consistent. See
-[DEPLOYMENT.md](DEPLOYMENT.md) for the full config reference, the `app_*` env
-vars, LDAPS/TLS, and backups.
-
-### 3. Bare metal on Debian/Ubuntu
-
-`install.sh` is an idempotent installer: it installs Node.js 20.x and OpenLDAP,
-configures the directory (modules, overlays, schema, the SSO groups), deploys
-the app to `/opt/sso-manager`, and creates a systemd unit.
-
-The only thing it requires is the LDAP admin password; the domain (base DN)
-defaults to `dc=example,dc=com` if you don't pass one:
-
-```bash
-sudo ./install.sh -p 'your-ldap-password' -b 'dc=yourdomain,dc=com'
-sudo systemctl enable --now sso-manager
-curl http://localhost:3001/health    # -> {"status":"ok"}
-```
-
-Run `sudo ./install.sh -h` for all flags (`-n` org name, `-o` port, `-j` JWT
-secret, `-s` SMTP, `--skip-ldap` to use an existing LDAP, `--dry-run`). Re-run
-it to update. Full details in [DEPLOYMENT.md](DEPLOYMENT.md) under *Method 2:
-Bare metal*.
+The `config/*-secrets.js` files are operator-edit seed artifacts (gitignored),
+not the authoritative store. For the full architecture, policies, token model,
+and rotation procedure, see theta-suite's
+**[Secrets docs](https://theta42.github.io/theta-suite/secrets.html)**.
 
 ## Architecture
 
@@ -135,7 +85,7 @@ Bare metal*.
        │ HTTP/HTTPS
        ▼
 ┌────────────────────────┐      ┌─────────────┐
-│  Express SSO Manager   │◄────►│   Redis     │
+│  Theta Directory       │◄────►│   Redis     │
 │  - OIDC provider       │      │ - sessions  │
 │  - web UI (:3001)      │      │ - models    │
 │  - management API      │      └─────────────┘
@@ -145,7 +95,7 @@ Bare metal*.
 ┌────────────────────────┐
 │  OpenLDAP (slapd)      │
 │  - users / groups      │
-│  - LDAPS :636          │─── legacy apps bind directly
+│  - LDAPS :636          │─── Linux hosts + LDAP apps bind directly
 │  - StartTLS :389       │
 └────────────────────────┘
 ```
@@ -158,26 +108,13 @@ required groups, LDAPS/TLS, direct-bind service accounts) live in:
 - [DEPLOYMENT.md](DEPLOYMENT.md) — Docker + bare metal, the config layers, the
   `app_*` env reference, LDAPS/TLS, backups, troubleshooting.
 - [API.md](API.md) — the management API.
-- [docs/](docs/) (GitHub Pages) — the same content broken into
-  [deployment](docs/deployment.md), [configuration](docs/configuration.md),
-  [OAuth/OIDC](docs/oauth.md), and [LDAP](docs/ldap.md).
+- [docs/](docs/) — the same content broken into
+  [OAuth/OIDC](docs/oauth.md) and [LDAP](docs/ldap.md), also published at the
+  unified [theta-suite docs site](https://theta42.github.io/theta-suite/sso/).
+- [CHANGELOG.md](CHANGELOG.md) — what changed in each release.
+- All of the above is also readable from the running app itself at `/docs` —
+  no internet access required.
 
-If you are pointing the app at your own existing LDAP server, see
-*LDAP requirements* in [DEPLOYMENT.md](DEPLOYMENT.md) — the directory needs the
-`pw-sha2`, `ppolicy`, `memberof`, and `refint` modules plus a small custom
-schema. The bundled Docker image and `install.sh` set all of that up for you.
-Required groups: `app_sso_admin` (full admin), `app_sso_oauth_admin` (manage
-OAuth clients only), `app_sso_invite` (invitation management) — see
-DEPLOYMENT.md for the full setup.
-
-## Development
-
-```bash
-cd nodejs
-npm install
-npm run dev      # nodemon auto-reload
-npm test         # jest test suite
-```
 
 ## License
 

config/proxy-secrets.js.example

@@ -0,0 +1,18 @@
+'use strict';
+
+// Example proxy secrets file. theta-env generates a real ./config/proxy-secrets.js
+// from this shape at setup (with empty clientId/clientSecret), then bootstrap.js
+// writes the SSO-generated OAuth client creds into it AND into OpenBao
+// (secret/proxy/conf). The proxy loads it via @simpleworkjs/conf, then overlays
+// secret/proxy/conf from OpenBao via @simpleworkjs/bao-conf at boot.
+//
+// The real file is gitignored (config/*-secrets.js) — never commit live creds.
+// This .example is tracked to document the expected shape only.
+module.exports = {
+	oidc: {
+		// The SSO registers the proxy as an OAuth client and writes the real
+		// values here (and into OpenBao). "set-me" is the bootstrap placeholder.
+		clientId: 'set-me',
+		clientSecret: 'set-me',
+	},
+};

directory_spec.md

@@ -1,8 +1,10 @@
 # Home-Lab Directory / Inventory — Design Spec
 
-Status: **Draft / agreed direction** (no code yet)
+Status: **Implemented** (v1.2.1+: model, admin API, UI; v1.3.x: automatic
+registration from theta-env + ldap-client). §9 adds the planned-consumer
+readiness review.
 Owner: wmantly
-Last updated: 2026-07-02
+Last updated: 2026-07-23
 
 ---
 
@@ -95,13 +97,24 @@ common query fields can be promoted to columns later.
 | column       | type        | notes |
 |--------------|-------------|-------|
 | `id`         | uuid / pk   | |
-| `kind`       | enum        | `proxmox_node` \| `container` \| `vm` \| `bare_metal` \| `service` |
+| `kind`       | enum        | `site` \| `host` \| `service` |
 | `name`       | text        | display name ("Home Assistant", "ct101") |
 | `slug`       | text unique | url-safe id used by the API |
 | `description`| text        | free text |
-| `metadata`   | jsonb       | `{ url, icon, fqdn, ip, port, tags[], … }` |
+| `metadata`   | jsonb       | `{ subType, ip, macAddress, address, vmid, port, externalPort, gitRepo, installPath, systemdService, os, kernel, isProduction, isExternalReachable, isPublic }` |
 | `created_at` / `updated_at` | timestamptz | |
 
+**Parent Enforcement Rules:**
+- A **Host** MUST have a parent **Site** or **Host**.
+- A **Service** MUST have a parent **Host**.
+- An **OAuth Integration** MUST have a parent **Service**.
+
+**LDAP Group Auto-Creation:**
+When a Host or Service is created, the system will automatically create two LDAP groups in the directory (if they do not already exist): 
+- `<slug>_access` (for standard user access)
+- `<slug>_admin` (for administrative access)
+Additional groups can still be linked manually.
+
 ### `resource_edge` — directed relationships (the graph)
 | column       | type   | notes |
 |--------------|--------|-------|
@@ -109,7 +122,7 @@ common query fields can be promoted to columns later.
 | `child_id`   | fk → resource | |
 | `relation`   | enum   | `runs_on` \| `hosts` \| `exposes` \| `depends_on` |
 
-Represents host←container←service (`hosts`/`runs_on`) and service→service
+Represents site←host←service (`hosts`/`runs_on`) and service→service
 (`depends_on`). Directed edges (not a single `parent_id` column) so a node can have
 multiple parents/children and multiple relation types.
 
@@ -164,12 +177,7 @@ Write endpoints (POST/PUT/DELETE) are **out of scope for v1**; population is man
 
 - **Interactive users:** existing session auth — `middleware.auth` validating the
   `auth-token` header (an `AuthToken`, `models/token.js`). No change.
-- **CI/CD (machine) access:** the app does **not yet** have a long-lived service
-  token — `AuthToken` is session-oriented. **Proposed small addition:** a
-  `ServiceToken` subclass in `models/token.js` (mirrors `AuthToken`/`ImpersonationToken`),
-  long-lived, read-only, passed in the same `auth-token` header. Track as its own
-  task; the discovery API should assume it exists but degrade to normal auth tokens
-  until then.
+- **CI/CD (machine) access:** scripts and external integrations (like jump hosts) will use the existing `ApiToken` system (Personal Access Tokens) passed in the `Authorization: Bearer sso_...` header. The `ApiToken` inherits the exact LDAP group permissions of the user who created it, seamlessly mapping to existing access controls.
 - **Read visibility (decision to confirm):** either (a) any authenticated user may
   read all resource metadata and only `/me` is filtered, or (b) list endpoints are
   themselves filtered to entitlement. Recommend **(a)** for a home lab — simpler,
@@ -195,7 +203,7 @@ Write endpoints (POST/PUT/DELETE) are **out of scope for v1**; population is man
 ## 7. Roadmap
 
 1. **v1 — Discovery API** (this spec's focus): SQL schema + migrations, read models,
-   `/api/discovery/*` endpoints, `ServiceToken` for CI/CD.
+   `/api/discovery/*` endpoints, `ApiToken` for CI/CD.
 2. **v2 — "My Access" dashboard**: swap `profile.ejs`'s static list for `/me`.
 3. **v3 — Admin CRUD UI**: manage resources/edges/group links (reusing `app.ui`
    widgets and the `oauth_clients.ejs` card+modal pattern); gated by
@@ -213,4 +221,185 @@ Write endpoints (POST/PUT/DELETE) are **out of scope for v1**; population is man
    `description` so LDAP-only external consumers see it? **Default: no** — keep LDAP
    for auth, SQL for inventory.
 4. **Read-visibility policy:** confirm option (a) vs (b) in §5.
-5. **Service token scope:** read-only globally, or per-token resource/kind scoping?
+5. **Service token scope:** Currently `ApiToken` shares the creator's full permissions. A future enhancement could scope tokens specifically to the Directory API.
+
+---
+
+## 9. Planned consumers — data-model & API readiness
+
+Five consumers the directory data should be able to power. None are being
+built yet; this section records what each needs, what already exists, and the
+gaps to close so the model/API never paints us into a corner.
+
+The recurring theme: **the graph model itself (Resource / ResourceEdge /
+ResourceGroup + LDAP groups) is sufficient for all five.** The gaps are
+(a) one new model (access requests), (b) machine-to-machine auth for the read
+API, (c) documented metadata conventions instead of new columns, and
+(d) change detection for the drift/sync consumers.
+
+### 9.1 End-user exploration ("Netflix-style" catalog + request access)
+
+A user browses everything that exists — part advertisement, part
+documentation — sees what they already have, and requests access to the rest.
+
+Already there:
+- `/api/discovery/me` (`getMyAccess`) — the "My Services" half.
+- `Resource.owner` + `<slug>_access` / `<slug>_admin` ResourceGroup links —
+  who approves, and which group an approval means joining.
+- The Notification model — the approval-request delivery mechanism.
+
+Gaps:
+1. **Catalog projection with metadata privacy.** `/api/discovery/resources`
+   returns full `metadata` to any authenticated user — including the OAuth
+   kind's `client_secret_hash`, and operator notes that may name internal
+   IPs. Needed: a per-kind public projection (name, description, kind,
+   subType, icon, address, hasAccess, requestable) and a private-key
+   convention for the rest (e.g. only `app_sso_directory_admin` sees full
+   metadata). This is a **fix worth doing before any catalog UI exists**.
+2. **`AccessRequest` model** — the one genuinely new model:
+   `{id, uid, resourceId, groupCn, status: pending|approved|denied, note,
+   requestedOn, decidedBy, decidedOn}`. Approval = LDAP group add + notify.
+   Endpoints: user POST/GET own; resource owner / directory admin
+   list/approve/deny.
+3. **Catalog metadata conventions**: `icon`, `tagline` (card-length blurb),
+   `requestable: false` for resources that shouldn't be advertised.
+
+### 9.2 SSH jump host (`username_-_{hostname-or-ip}@publicHost`)
+
+A public jump host parses the target out of the SSH username, checks the user
+may reach that host, and proxies the connection (WinSCP-friendly: one
+username string, no interactive menu needed — though an interactive picker on
+plain `username@` login is the same query).
+
+Already there:
+- Hosts carry `ip` (and `host_<hostname>` slugs to resolve by name).
+- Access is already group-based (`<slug>_access`), checkable via LDAP alone —
+  the jump host can run entirely off LDAP (SSSD) + one directory query.
+- User SSH keys are in LDAP (openssh-lpk) — the jump host authenticates the
+  real user without local accounts.
+
+Gaps:
+1. **Machine auth for the access query.** The jump host must ask "may user X
+   reach host Y" / "list hosts user X may reach" *about another user*.
+   `getMyAccess` only answers for the calling user. Needed: a
+   service-token-authenticated endpoint (`GET
+   /api/discovery/access/:uid[/:slug]`). `ServiceToken` already exists and
+   is even linked to a resource (`resource_id`) — what's missing is an auth
+   middleware that accepts it and a permission rule ("service tokens may
+   read access info, scoped read-only").
+2. **Connection metadata conventions** on hosts: `sshPort` (default 22),
+   optional `fqdn` (when IP is dynamic), optional `jumpVia` edge relation if
+   multi-hop topologies ever appear.
+3. Document the username grammar (`{uid}_-_{host-slug-or-ip}`) here so the
+   seed/ldap-client keep host slugs DNS-safe (they already are: slugify
+   strips everything but `[a-z0-9-]`).
+
+### 9.3 Firewall port-forward rules (build / update / drift-test)
+
+An automation renders the public firewall's forwarding table from the
+directory, applies it, and alerts on drift in either direction.
+
+Already there:
+- `metadata.port` / `metadata.externalPort` / `metadata.ip` /
+  `metadata.isExternalReachable` — the core mapping data, already seeded for
+  the stack's own services.
+
+Gaps:
+1. **Port-mapping convention is too thin for real rules**: no protocol, no
+   multi-port services. Adopt `metadata.portMappings: [{proto: "tcp"|"udp",
+   external: n, internal: n, comment}]` as the authoritative form
+   (`port`/`externalPort` stay as the simple single-mapping case).
+2. **Drift detection needs cheap change polling**: an `updated_on` timestamp
+   on resources surfaced in the graph API, or a graph-level etag/hash, so
+   the runner can poll without diffing full payloads. (The ORM already
+   publishes create/update events internally — a future push feed can ride
+   that; polling comes first.)
+3. Same **service-token read auth** as 9.2 — automation must not run on a
+   human's session token.
+
+### 9.4 Local DNS / mDNS
+
+A DNS (or mDNS advertiser) zone is generated from the directory: hosts get
+A records from `metadata.ip`, services get CNAMEs/records from their
+addresses, sites map to zones.
+
+Already there:
+- `host_<hostname>` + `ip` covers A records; `site_<name>` is a natural zone
+  boundary; service `address` yields names.
+
+Gaps:
+1. **Name conventions**: `metadata.dnsNames: []` for extra aliases, and a
+   documented rule for which name wins (slug vs `address` hostname). TTL
+   only if someone actually needs per-record TTLs — default is fine.
+2. Same **change detection** as 9.3 (poll `updated_on` / etag; push later).
+3. Nothing else — this consumer is nearly free once 9.3's conventions land.
+
+### 9.5 Access control for hosts
+
+Who may log in to / sudo on which machine, driven by the directory.
+
+Already there — this is the original point of the system:
+- `<slug>_access` / `<slug>_admin` groups are auto-provisioned per host;
+  ldap-client configures SSSD/PAM against the directory; `sudoRole` and
+  openssh-lpk schemas cover sudo and SSH keys.
+
+Gaps:
+1. **Close the loop in ldap-client**: joined hosts should set an SSSD access
+   filter (`access_provider = ldap`, filter on `host_<hostname>_access`
+   membership) so directory group membership *is* login permission, not just
+   identity. Today the registration exists but enforcement is host-side
+   convention.
+2. **`accessLevel` granularity**: ResourceGroup's `member`/`owner` maps to
+   login/admin today; if finer roles emerge (e.g. `login` vs `sudo` vs
+   `admin`), extend the enum — the join-table shape already supports it.
+
+### 9.6 Consolidated work list (model/API only, no consumers)
+
+Ordered by how much they unblock:
+
+1. **Metadata privacy projection** on the read API (blocks 9.1; fixes the
+   `client_secret_hash` exposure regardless of any consumer).
+2. **Service-token auth for `/api/discovery/*`** + `access/:uid` endpoint
+   (blocks 9.2, 9.3; ServiceToken model already exists).
+3. **`AccessRequest` model + endpoints** (blocks 9.1's request half).
+4. **Metadata conventions doc entries** (`sshPort`, `portMappings`,
+   `dnsNames`, `icon`, `tagline`, `requestable`) in `docs/directory.md` —
+   conventions, not schema changes; the json column already holds them.
+5. **`updated_on` in graph output / graph etag** (blocks drift/DNS
+   freshness; trivial once surfaced).
+
+---
+
+## 10. Subtype Management & Metrics Drivers Architecture
+
+The Directory incorporates a **4-tier Driver Resolution Engine** (`services/driver_registry.js`) that binds resource `subType` metadata to specific telemetry, log streaming, and operational management protocols.
+
+### Subtype Matrix & Drivers
+
+| Subtype Category | Supported Subtypes | Primary Driver | Management Capabilities | Telemetry & Metrics |
+| :--- | :--- | :--- | :--- | :--- |
+| **Service Managers** | `systemd`, `openrc`, `windows_service` | `ThetaAgentDriver` / Systemd | `start`, `stop`, `restart`, `reload` | CPU, Memory, Active PID, SubState |
+| **Containers & Stacks** | `docker`, `docker_compose` | `DockerSocketDriver` / Agent | `start`, `stop`, `restart`, `pause` | CPU %, Memory Limit/Usage, Net/Block I/O |
+| **Virtualization & Hypervisors** | `proxmox`, `lxc`, `kvm`, `esxi`, `libvirt_kvm`, `vps_generic` | `ProxmoxDriver` / Hypervisor | `start`, `stop`, `shutdown`, `reboot` | Guest VMID CPU/RAM/Disk, Parent Hypervisor status |
+| **Networking & Appliances** | `wireguard`, `unifi_ap`, `unifi_switch`, `pfsense` | `NetworkDriver` | `restart`, `locate`, `sync` | Connected Clients, Handshakes, Gateway RTT, Channels |
+| **Databases & Vaults** | `postgresql`, `redis`, `openbao_vault` | `DbDriver` | `flush`, `seal`, `unseal` | DB Size, Connections, Hit Rates, Active Leases |
+| **Orchestration** | `k8s_pod`, `k8s_deployment` | `K8sDriver` | `scale`, `restart`, `rollout_restart` | Desired/Ready Replicas, Pod Phase, IP |
+| **Workstations** | `desktop_linux`, `desktop_windows` | `ThetaAgentDriver` | `reboot`, `shutdown`, Display Manager | CPU, Memory, GPU, Active Sessions |
+
+*(Note: Reverse Proxy subtypes like Nginx/HAProxy/Caddy/Traefik are excluded per environment configuration).*
+
+### 4-Tier Driver Resolution Engine
+1. **Direct Agent Execution**: If `theta-agent` is connected directly to the target resource.
+2. **Subtype-Specific Driver**: Executes specialized protocol driver (e.g. Proxmox API, Docker Engine API, DB Driver).
+3. **Ancestor / Hypervisor Fallback**: If an LXC/KVM guest lacks a direct agent, queries its parent Proxmox hypervisor node for metrics and power controls.
+4. **Unmanaged Fallback**: Reports unmanaged status cleanly without breaking UI/API contracts.
+
+### Subtype Operations Endpoints
+- `GET /api/directory-admin/resources/:id/driver-metrics` — Real-time telemetry payload
+- `POST /api/directory-admin/resources/:id/driver-action` — Execute management action (`{ action, params }`)
+- `GET /api/directory-admin/resources/:id/driver-logs` — Tail log output (`?lines=100`)
+
+## 11. Multi-Site
+
+This directory can run across multiple sites (one **master** with write authority, any number of **spoke** read-only replicas that stay live-synced after joining), coordinate master promotion, and share the agent-signing key across sites. Full design and operational detail: [`docs/site-join.md`](docs/site-join.md) and, at the suite level, `theta-suite`'s `docs/MULTI_SITE_SPEC.md`.
+

docker-compose.e2e.yml

@@ -0,0 +1,93 @@
+# End-to-end test of the LDAP byte-pump tunnel (DESIGN.md §4).
+#
+# Spins up OpenLDAP + Redis, a real SSO server (bin/www, so the WSS relay is
+# live), and a client that simulates the agent: it enrolls one, connects over
+# WSS, sends a real LDAP bind as raw bytes, and verifies the SSO relays it into
+# OpenLDAP and pipes the response back.
+#
+#   docker compose -f docker-compose.e2e.yml up --build --abort-on-container-exit
+#   # exit code 0 = tunnel works; the client prints E2E PASS.
+
+services:
+  ldap:
+    build:
+      context: .
+      dockerfile: Dockerfile.openldap
+    environment:
+      - LDAP_BASE_DN=dc=test,dc=local
+      - LDAP_ADMIN_PASS=secret
+      - ORG_NAME=Test SSO
+    command: ["sleep", "infinity"]
+    healthcheck:
+      test: ["CMD-SHELL", "ldapsearch -x -H ldap://localhost:389 -b '' -s base '(objectClass=*)' >/dev/null 2>&1"]
+      interval: 2s
+      timeout: 3s
+      retries: 20
+      start_period: 5s
+    volumes:
+      - ldap-data:/var/lib/ldap
+      - ldap-certs:/etc/openldap/certs
+
+  redis:
+    image: redis:7-alpine
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 2s
+      timeout: 3s
+      retries: 15
+
+  sso:
+    build:
+      context: .
+      dockerfile: Dockerfile.test-runner
+    command: ["node", "bin/www"]
+    environment:
+      - NODE_ENV=test
+      - NODE_PORT=3001
+      # Test OpenBao (theta-test-bao) — sso-broker token so the SSO can sign
+      # high-risk agent commands and read node-scoped secrets.
+      - VAULT_ADDR=http://theta-test-bao:8200
+      - VAULT_TOKEN=${VAULT_TOKEN:-}
+      - app_ldap__url=ldap://ldap:389
+      - app_ldap__bindDN=cn=admin,dc=test,dc=local
+      - app_ldap__bindPassword=secret
+      - app_ldap__userBase=ou=people,dc=test,dc=local
+      - app_ldap__groupBase=ou=groups,dc=test,dc=local
+      - app_redis__redisConf__url=redis://redis:6379
+      - REDIS_URL=redis://redis:6379
+      - app_oauth__jwtSecret=test-jwt-secret-for-testing-only
+      - app_name=Test SSO
+    depends_on:
+      ldap:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+  client:
+    build:
+      context: .
+      dockerfile: Dockerfile.test-runner
+    command: ["sh", "-c", "seed-test-user && node test/tunnel_e2e.js"]
+    environment:
+      - NODE_ENV=test
+      - SSO_URL=http://sso:3001
+      - app_ldap__url=ldap://ldap:389
+      - app_ldap__bindDN=cn=admin,dc=test,dc=local
+      - app_ldap__bindPassword=secret
+      - app_ldap__userBase=ou=people,dc=test,dc=local
+      - app_ldap__groupBase=ou=groups,dc=test,dc=local
+      - app_redis__redisConf__url=redis://redis:6379
+      - REDIS_URL=redis://redis:6379
+      - app_oauth__jwtSecret=test-jwt-secret-for-testing-only
+      - app_name=Test SSO
+    depends_on:
+      sso:
+        condition: service_started
+      ldap:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+volumes:
+  ldap-data:
+  ldap-certs:

docker-compose.multisite-e2e.yml

@@ -0,0 +1,69 @@
+# End-to-end test of the real, shipped multi-site join flow (docs/site-join.md).
+#
+# Spins up two full all-in-one instances (app + bundled slapd each, like
+# docker-compose.repl-test.yml) — "master" and "spoke" — plus a client that
+# drives the actual HTTP API a human/operator would use: mint a site join key
+# on master, join from spoke, verify the spoke adopted the catalog, went
+# read-only, and reports live WAN health.
+#
+#   docker compose -f docker-compose.multisite-e2e.yml up --build --abort-on-container-exit
+#   # exit code 0 = MULTISITE E2E PASS
+#
+# slapcat (used by POST /api/site/export) only sees the LDAP data of the
+# container it runs in, so this MUST use the all-in-one image (master and
+# spoke each carry their own slapd) — the split ldap+redis+app harness used
+# by docker-compose.test.yml/e2e.yml won't exercise export/join at all.
+
+services:
+  master:
+    build:
+      context: .
+      dockerfile: Dockerfile.openldap
+    container_name: multisite_e2e_master
+    environment:
+      - LDAP_BASE_DN=dc=master,dc=test
+      - LDAP_ADMIN_PASS=secret
+      - ORG_NAME=E2E Master
+      - app_oauth__jwtSecret=e2e-multisite-master-jwt-secret
+    healthcheck:
+      test: ["CMD-SHELL", "wget -qO- http://localhost:3001/health >/dev/null 2>&1"]
+      interval: 2s
+      timeout: 3s
+      retries: 40
+      start_period: 5s
+
+  spoke:
+    build:
+      context: .
+      dockerfile: Dockerfile.openldap
+    container_name: multisite_e2e_spoke
+    environment:
+      - LDAP_BASE_DN=dc=spoke,dc=test
+      - LDAP_ADMIN_PASS=secret
+      - ORG_NAME=E2E Spoke
+      - app_oauth__jwtSecret=e2e-multisite-spoke-jwt-secret
+    healthcheck:
+      test: ["CMD-SHELL", "wget -qO- http://localhost:3001/health >/dev/null 2>&1"]
+      interval: 2s
+      timeout: 3s
+      retries: 40
+      start_period: 5s
+
+  client:
+    build:
+      context: .
+      dockerfile: Dockerfile.test-runner
+    command: ["sh", "-c", "node test/multisite_join_e2e.js"]
+    environment:
+      - MASTER_URL=http://master:3001
+      - SPOKE_URL=http://spoke:3001
+      - MASTER_LDAP_HOST=master
+      - MASTER_BASE_DN=dc=master,dc=test
+      - SPOKE_LDAP_HOST=spoke
+      - SPOKE_BASE_DN=dc=spoke,dc=test
+      - LDAP_ADMIN_PASS=secret
+    depends_on:
+      master:
+        condition: service_healthy
+      spoke:
+        condition: service_healthy

docker-compose.repl-test.yml

@@ -0,0 +1,36 @@
+services:
+  site1:
+    build: .
+    container_name: sso_site1
+    environment:
+      - LDAP_SERVER_ID=1
+      - LDAP_REPLICATION_HOSTS=ldap://site2:389
+      - LDAP_BASE_DN=dc=test,dc=local
+      - LDAP_ADMIN_PASS=secret
+    ports:
+      - "3001:3001"
+      - "10389:389"
+    volumes:
+      - site1-ldap:/var/lib/ldap
+      - site1-redis:/data
+
+  site2:
+    build: .
+    container_name: sso_site2
+    environment:
+      - LDAP_SERVER_ID=2
+      - LDAP_REPLICATION_HOSTS=ldap://site1:389
+      - LDAP_BASE_DN=dc=test,dc=local
+      - LDAP_ADMIN_PASS=secret
+    ports:
+      - "3002:3001"
+      - "20389:389"
+    volumes:
+      - site2-ldap:/var/lib/ldap
+      - site2-redis:/data
+
+volumes:
+  site1-ldap:
+  site1-redis:
+  site2-ldap:
+  site2-redis:

docker-compose.test.yml

@@ -0,0 +1,81 @@
+# Docker Compose for running the SSO Manager test suite.
+#
+# Spins up:
+#   ldap   — OpenLDAP + Redis (all-in-one image, slapd + redis only, no app)
+#   redis  — Standalone Redis for the app's model/token storage
+#   test-runner — Seeds the test user, then runs `npm test`
+#
+# Usage:
+#   docker compose -f docker-compose.test.yml up --build
+#   # Or to run a specific test file:
+#   docker compose -f docker-compose.test.yml run --rm test-runner npx jest tests/auth.test.js
+#
+# The LDAP service uses the same Dockerfile.openldap image as production but
+# overrides the command to only start slapd + redis (the entrypoint handles
+# slapd.conf generation, directory initialization, and Redis startup before
+# running the given command — "sleep infinity" keeps it alive).
+#
+# The test-runner connects to ldap:389 and redis:6379 via Docker networking.
+# app_* env vars override conf/secrets.js (highest precedence in
+# @simpleworkjs/conf), so the production secrets.js is never read.
+
+services:
+  ldap:
+    build:
+      context: .
+      dockerfile: Dockerfile.openldap
+    environment:
+      - LDAP_BASE_DN=dc=test,dc=local
+      - LDAP_ADMIN_PASS=secret
+      - ORG_NAME=Test SSO
+    # The entrypoint starts slapd + redis, then runs whatever command is given.
+    # "sleep infinity" keeps the container alive so the test-runner can connect.
+    command: ["sleep", "infinity"]
+    healthcheck:
+      test: ["CMD-SHELL", "ldapsearch -x -H ldap://localhost:389 -b '' -s base '(objectClass=*)' >/dev/null 2>&1"]
+      interval: 2s
+      timeout: 3s
+      retries: 20
+      start_period: 5s
+    volumes:
+      - ldap-data:/var/lib/ldap
+      - ldap-certs:/etc/openldap/certs
+
+  redis:
+    image: redis:7-alpine
+    healthcheck:
+      test: ["CMD", "redis-cli", "ping"]
+      interval: 2s
+      timeout: 3s
+      retries: 15
+
+  test-runner:
+    build:
+      context: .
+      dockerfile: Dockerfile.test-runner
+    environment:
+      # Tell the app which environment it's in (loads conf/test.js for Redis prefix)
+      - NODE_ENV=test
+      # LDAP — point at the ldap service container
+      - app_ldap__url=ldap://ldap:389
+      - app_ldap__bindDN=cn=admin,dc=test,dc=local
+      - app_ldap__bindPassword=secret
+      - app_ldap__userBase=ou=people,dc=test,dc=local
+      - app_ldap__groupBase=ou=groups,dc=test,dc=local
+      # Redis — point at the redis service container
+      - app_redis__redisConf__url=redis://redis:6379
+      # Also used by tests/globalSetup.js (direct Redis client, not @simpleworkjs/conf)
+      - REDIS_URL=redis://redis:6379
+      # JWT secret (required by the app, not sensitive in test)
+      - app_oauth__jwtSecret=test-jwt-secret-for-testing-only
+      # App name
+      - app_name=Test SSO
+    depends_on:
+      ldap:
+        condition: service_healthy
+      redis:
+        condition: service_healthy
+
+volumes:
+  ldap-data:
+  ldap-certs:

docker-entrypoint.sh

@@ -6,12 +6,12 @@
 # production, run a dedicated LDAP server and point the app at it via app_*
 # env vars (or a mounted conf/secrets.js) using the app-only image.
 #
-# The app reads its configuration from conf/base.js + conf/secrets.js, deep-merged
-# by @simpleworkjs/conf, with `app_*` environment variables as the
-# highest-precedence override layer. This entrypoint exports those `app_*`
-# vars so the app connects to the bundled slapd without any mounted secrets
-# file. Any `app_*` var already set in the environment wins (the values below
-# are defaults/fallbacks only).
+# The app reads its configuration from conf/base.js + a secrets file, deep-merged
+# by @simpleworkjs/conf (requires >= 1.2.0, pinned in nodejs/package-lock.json),
+# with `app_*` environment variables as the highest-precedence override layer.
+# This entrypoint exports those `app_*` vars so the app connects to the bundled
+# slapd without any mounted secrets file. Any `app_*` var already set in the
+# environment wins (the values below are defaults/fallbacks only).
 
 set -e
 
@@ -33,14 +33,15 @@ error() { echo "[ERROR] $*" >&2; }
 # ── Optional: load operational config from a mounted secrets.js ──────────────
 # The unified theta-env stack mounts ./config/sso-secrets.js at /config and
 # treats it as the authoritative source for the SSO's config (LDAP base, admin
-# password, org name, JWT secret, ...). When present, symlink it into
-# /app/conf/secrets.js so @simpleworkjs/conf reads it, and override the
-# env-derived operational vars below with the file's values. When absent
-# (standalone / env-var deployments) the env vars set above stay in effect and
-# the app_* exports further down are emitted as before.
+# password, org name, JWT secret, ...). When present, point CONF_SECRETS at it
+# so @simpleworkjs/conf reads it directly (no write access to /app/conf
+# needed), and override the env-derived operational vars below with the
+# file's values. When absent (standalone / env-var deployments) the env vars
+# set above stay in effect and the app_* exports further down are emitted as
+# before.
 SECRETS_JS_MODE=0
 if [[ -f /config/sso-secrets.js ]]; then
-    ln -sf /config/sso-secrets.js /app/conf/secrets.js
+    export CONF_SECRETS=/config/sso-secrets.js
     SECRETS_JS_MODE=1
     # Pull the entrypoint's operational vars out of secrets.js in one node call.
     # Node emits `KEY<TAB>base64(value)` lines; we decode each with base64 -d and
@@ -75,11 +76,22 @@ fi
 # ── Locate the OpenLDAP module directory ────────────────────────────────────
 # slapd.conf needs `modulepath` to find pw-sha2/ppolicy/memberof/refint. The
 # path varies by distro; auto-detect rather than hardcode.
+# /opt/openldap/libexec/openldap is first: that is the from-source build (see
+# Dockerfile.openldap), which is the only one carrying the nestgroup overlay.
 MODULE_PATH=""
-for p in /usr/lib/openldap /usr/lib/ldap /usr/local/lib/openldap /opt/local/lib/openldap; do
+for p in /opt/openldap/libexec/openldap /usr/lib/openldap /usr/lib/ldap /usr/local/lib/openldap /opt/local/lib/openldap; do
     if [[ -d "$p" ]]; then MODULE_PATH="$p"; break; fi
 done
 
+# Nested-group support is only available when slapd was built with the
+# nestgroup overlay. Detect rather than assume, so this entrypoint still
+# produces a working slapd.conf against a distro OpenLDAP (where the app falls
+# back to resolving nesting itself -- see nodejs/models/group_ldap.js).
+NESTGROUP_AVAILABLE=0
+if [[ -n "$MODULE_PATH" && -f "$MODULE_PATH/nestgroup.so" ]]; then
+    NESTGROUP_AVAILABLE=1
+fi
+
 # ── TLS certificate for LDAPS / StartTLS ────────────────────────────────────
 # Legacy apps (e.g. the theta42/proxy, Gitea, Emby) bind to LDAP directly over the
 # network. To keep password binds off the wire in cleartext we expose LDAPS
@@ -124,6 +136,8 @@ include         /etc/openldap/schema/theta42.schema
 include         /etc/openldap/schema/sudo.schema
 include         /etc/openldap/schema/openssh-lpk.schema
 
+SERVER_ID_PLACEHOLDER
+
 # Module loading (pw-sha2 provides {SSHA512} used by the app for user passwords;
 # ppolicy/memberof/refint are the overlays the app depends on). On OpenLDAP 2.5+
 # the ppolicy schema (pwdPolicy, pwdAccountLockedTime, ...) is built into
@@ -136,6 +150,9 @@ moduleload      pw-sha2
 moduleload      ppolicy
 moduleload      memberof
 moduleload      refint
+moduleload      auditlog
+NESTGROUP_MODULE_PLACEHOLDER
+SYNCPROV_MODULE_PLACEHOLDER
 
 # TLS (LDAPS on 636 + StartTLS on 389). Cert/key paths are fixed; the files are
 # generated/mounted above. We accept clients without their own cert (the common
@@ -183,6 +200,14 @@ memberof-memberof-ad memberOf
 overlay         refint
 refint_attributes memberOf member manager owner
 
+NESTGROUP_OVERLAY_PLACEHOLDER
+
+# auditlog overlay (LDIF audit trail of all changes)
+overlay         auditlog
+auditlog        /var/lib/ldap/auditlog.ldif
+
+REPLICATION_BLOCK_PLACEHOLDER
+
 # Access controls
 access to attrs=userPassword
     by dn="BIND_DN_PLACEHOLDER" write
@@ -210,6 +235,61 @@ else
     sed -i "/^SLAPMODULEPATH$/d" /etc/openldap/slapd.conf
 fi
 
+# ── Nested groups (nestgroup overlay) ──
+# Three of the four flags, deliberately:
+#
+#   member-filter     (member=X) finds parent groups transitively. This is what
+#                     Group.list(dn) rides on -- the core access question.
+#   memberof-filter   (memberOf=X) matches members of nested groups. This is
+#                     what SSSD's ldap_access_filter uses, so SSH/sudo inherit
+#                     nesting without any client-side walking.
+#   memberof-values   expands memberOf when reading a user, so anything that
+#                     reads the attribute rather than searching still sees the
+#                     full picture.
+#
+# member-values is deliberately NOT enabled. It expands the `member` attribute
+# when reading a *group*, which sounds symmetric but destroys the distinction
+# between "listed on this group" and "reachable through a nested one" -- and
+# that distinction is not recoverable afterwards, because the raw values are
+# simply not returned. The Groups UI needs it to show nested groups as nested
+# rather than as a crowd of phantom users, and un-nesting needs it to know what
+# it is actually removing. Transitive *answers* come from the filter flags and
+# from Group.effectiveMembers(), which computes the closure explicitly.
+if [[ "$NESTGROUP_AVAILABLE" == "1" ]]; then
+    info "nestgroup overlay available — nested groups resolved server-side"
+    sed -i "s|^NESTGROUP_MODULE_PLACEHOLDER$|moduleload      nestgroup|" /etc/openldap/slapd.conf
+    NESTGROUP_BLOCK="# nestgroup overlay (server-side nested group evaluation)\noverlay         nestgroup\nnestgroup-base  ou=groups,${LDAP_BASE_DN}\nnestgroup-flags member-filter memberof-filter memberof-values"
+    sed -i "s|^NESTGROUP_OVERLAY_PLACEHOLDER$|${NESTGROUP_BLOCK}|" /etc/openldap/slapd.conf
+else
+    info "nestgroup overlay not present in ${MODULE_PATH:-<no module path>} — nested groups will be resolved by the app instead"
+    sed -i "/^NESTGROUP_MODULE_PLACEHOLDER$/d" /etc/openldap/slapd.conf
+    sed -i "/^NESTGROUP_OVERLAY_PLACEHOLDER$/d" /etc/openldap/slapd.conf
+fi
+
+# ── Multi-Master Replication Configuration ──
+if [[ -n "${LDAP_SERVER_ID:-}" && -n "${LDAP_REPLICATION_HOSTS:-}" ]]; then
+    info "Configuring Multi-Master replication (Server ID: ${LDAP_SERVER_ID})"
+    sed -i "s|^SERVER_ID_PLACEHOLDER|ServerID ${LDAP_SERVER_ID}|" /etc/openldap/slapd.conf
+    sed -i "s|^SYNCPROV_MODULE_PLACEHOLDER|moduleload      syncprov|" /etc/openldap/slapd.conf
+    
+    # Generate syncrepl blocks
+    REPL_BLOCK="overlay syncprov\nsyncprov-checkpoint 100 10\nsyncprov-sessionlog 100\n\n"
+    RID=100
+    for HOST in ${LDAP_REPLICATION_HOSTS}; do
+        RID=$((RID + 1))
+        REPL_BLOCK="${REPL_BLOCK}syncrepl rid=${RID}\n  provider=${HOST}\n  type=refreshAndPersist\n  retry=\"60 +\"\n  searchbase=\"${LDAP_BASE_DN}\"\n  bindmethod=simple\n  binddn=\"${LDAP_BIND_DN}\"\n  credentials=\"${LDAP_ADMIN_PASS}\"\n\n"
+    done
+    REPL_BLOCK="${REPL_BLOCK}mirrormode on\n"
+    
+    # Replace placeholder (awk is safer for multiline replacements than sed)
+    awk -v repl="$(printf '%b' "$REPL_BLOCK")" '{gsub(/REPLICATION_BLOCK_PLACEHOLDER/, repl)}1' /etc/openldap/slapd.conf > /etc/openldap/slapd.conf.tmp
+    mv /etc/openldap/slapd.conf.tmp /etc/openldap/slapd.conf
+else
+    sed -i "/^SERVER_ID_PLACEHOLDER/d" /etc/openldap/slapd.conf
+    sed -i "/^SYNCPROV_MODULE_PLACEHOLDER/d" /etc/openldap/slapd.conf
+    sed -i "/^REPLICATION_BLOCK_PLACEHOLDER/d" /etc/openldap/slapd.conf
+fi
+
 chown ldap:ldap /etc/openldap/slapd.conf 2>/dev/null || true
 chown -R ldap:ldap /var/lib/ldap 2>/dev/null || true
 
@@ -219,7 +299,7 @@ info "Starting OpenLDAP (base DN: ${LDAP_BASE_DN})..."
 # -h listens on ldap:///  (389: plain + StartTLS) and ldaps:/// (636: LDAPS).
 # ldapi:/// is intentionally omitted: its default socket dir doesn't exist on
 # Alpine and the container only uses simple bind over ldap://localhost:389.
-slapd -d 0 -u ldap -g ldap -f /etc/openldap/slapd.conf -h "ldap:/// ldaps:///" &
+slapd -d 256 -u ldap -g ldap -f /etc/openldap/slapd.conf -h "ldap:/// ldaps:///" >> /var/lib/ldap/slapd.log 2>&1 &
 SLAPD_PID=$!
 
 # Wait for slapd to answer the root DSE (means it's up, regardless of DB state).
@@ -267,7 +347,7 @@ objectClass: organizationalRole
 objectClass: pwdPolicy
 cn: ppolicy
 pwdAttribute: 2.5.4.35
-pwdLockout: FALSE
+pwdLockout: TRUE
 pwdMustChange: FALSE
 pwdAllowUserChange: TRUE
 EOF
@@ -275,7 +355,11 @@ EOF
     # Required SSO groups. The app gates admin/invite/oauth-admin on these;
     # app_sso_service_account is a marker (not a permission gate) for
     # non-person accounts -- see the Users page.
-    for group in app_sso_admin app_sso_invite app_sso_oauth_admin app_sso_service_account; do
+    #
+    # god_admin is the global super group (docs/GROUPS.md §2), the top of the
+    # group-inheritance lattice. It is seeded here so it exists from first boot;
+    # the theta-suite bootstrap puts the first admin person into it.
+    for group in god_admin app_sso_admin app_sso_invite app_sso_oauth_admin app_sso_service_account; do
         ldapadd -x -D "$LDAP_BIND_DN" -w "$LDAP_ADMIN_PASS" -H ldap://localhost:389 << EOF || true
 dn: cn=${group},ou=groups,${LDAP_BASE_DN}
 objectClass: groupOfNames
@@ -286,6 +370,27 @@ member: ${LDAP_BIND_DN}
 EOF
     done
 
+    # Nest god_admin into the SSO admin groups, so god admins hold those rights
+    # by membership rather than by a special case in app code. This is what makes
+    # the privilege visible to every consumer -- SSSD, sudo, anything binding
+    # LDAP directly -- instead of only to callers that happen to route through
+    # utils/permission.js.
+    #
+    # app_sso_service_account is deliberately excluded: it is a marker for
+    # non-person accounts, not a permission, and nesting admins into it would
+    # misclassify them as service accounts on the Users page.
+    if [[ "$NESTGROUP_AVAILABLE" == "1" ]]; then
+        for group in app_sso_admin app_sso_invite app_sso_oauth_admin; do
+            ldapmodify -x -D "$LDAP_BIND_DN" -w "$LDAP_ADMIN_PASS" -H ldap://localhost:389 >/dev/null 2>&1 << EOF || true
+dn: cn=${group},ou=groups,${LDAP_BASE_DN}
+changetype: modify
+add: member
+member: cn=god_admin,ou=groups,${LDAP_BASE_DN}
+EOF
+        done
+        info "Nested god_admin into the SSO admin groups"
+    fi
+
     info "LDAP directory initialized"
 else
     info "LDAP directory already initialized — skipping seed"
@@ -345,6 +450,14 @@ if [[ "${SECRETS_JS_MODE:-0}" != 1 ]]; then
     export app_ldap__bindPassword="${app_ldap__bindPassword:-$LDAP_ADMIN_PASS}"
     export app_ldap__userBase="${app_ldap__userBase:-ou=people,${LDAP_BASE_DN}}"
     export app_ldap__groupBase="${app_ldap__groupBase:-ou=groups,${LDAP_BASE_DN}}"
+    # Tell the app whether slapd resolves nested groups for it. When true the app
+    # trusts a plain (member=) search to be transitive; when false it computes
+    # the closure itself. Getting this wrong in the "true" direction silently
+    # under-grants, so it is derived from the same nestgroup.so probe that
+    # decides whether the overlay is configured at all -- never hardcoded.
+    if [[ "$NESTGROUP_AVAILABLE" == "1" ]]; then
+        export app_ldap__nestedGroupsServerSide="${app_ldap__nestedGroupsServerSide:-true}"
+    fi
     export app_oauth__jwtSecret="${app_oauth__jwtSecret:-$JWT_SECRET}"
     # OIDC issuer advertised in /.well-known/openid-configuration. Default to the
     # public https URL on the SSO subdomain of the LDAP domain; override with

docs/_config.yml

@@ -1,9 +1,55 @@
 title: SSO Manager
-description: A self-hosted OpenID Connect provider with an OpenLDAP directory and a web management UI
-theme: jekyll-theme-cayman
-show_downloads: false
+description: A self-hosted OpenID Connect provider with a bundled OpenLDAP directory and a web management UI, for home labs and small businesses that want their own identity provider.
+url: "https://theta42.github.io"
+baseurl: "/sso-manager-node"
+logo: /assets/img/theta42.svg
+lang: en_US
+
+plugins:
+  - jekyll-seo-tag
+  - jekyll-sitemap
+
 github:
   repository_url: https://github.com/theta42/sso-manager-node
   zip_url: https://github.com/theta42/sso-manager-node/archive/refs/heads/master.zip
   tar_url: https://github.com/theta42/sso-manager-node/archive/refs/heads/master.tar.gz
-  repository_name: theta42/sso-manager-node
+  repository_name: theta42/sso-manager-node
+
+nav:
+  - title: Home
+    page: /
+    icon: fa-house
+  - title: Deployment
+    page: /deployment.html
+    icon: fa-server
+  - title: Configuration
+    page: /configuration.html
+    icon: fa-gears
+  - title: OAuth
+    page: /oauth.html
+    icon: fa-key
+  - title: LDAP
+    page: /ldap.html
+    icon: fa-address-book
+  - title: Directory
+    page: /directory.html
+    icon: fa-server
+  - title: Plugins
+    page: /plugins.html
+    icon: fa-plug
+  # API.md lives at the repo root, not under docs/, so Jekyll never renders an
+  # api.html for it — link the source directly, same as the Changelog.
+  - title: API
+    url: https://github.com/theta42/sso-manager-node/blob/master/API.md
+    icon: fa-code
+  - title: Changelog
+    url: https://github.com/theta42/sso-manager-node/blob/master/CHANGELOG.md
+    icon: fa-list
+
+defaults:
+  - scope:
+      path: ""
+      type: "pages"
+    values:
+      layout: default
+      image: /assets/img/theta42.svg

docs/_layouts/default.html

@@ -0,0 +1,82 @@
+<!doctype html>
+<html lang="en">
+<head>
+	<meta charset="utf-8">
+	<meta name="viewport" content="width=device-width, initial-scale=1, shrink-to-fit=no">
+	<link rel="icon" type="image/svg+xml" href="{{ '/assets/img/theta42.svg' | relative_url }}">
+
+	{% seo title=false %}
+	<title>{% if page.title %}{{ page.title }} &middot; {% endif %}{{ site.title }}</title>
+
+	<link rel="stylesheet" href="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/css/bootstrap.min.css">
+	<link rel="stylesheet" href="https://cdnjs.cloudflare.com/ajax/libs/font-awesome/6.5.2/css/all.min.css">
+	<link rel="stylesheet" href="{{ '/assets/css/style.css' | relative_url }}">
+</head>
+<body class="d-flex flex-column min-vh-100">
+
+	<nav class="navbar navbar-expand-md navbar-dark bg-dark fixed-top">
+		<div class="container-fluid px-3">
+			<a class="navbar-brand d-flex align-items-center" href="{{ '/' | relative_url }}">
+				<img src="{{ '/assets/img/theta42.svg' | relative_url }}" height="28" class="me-2" alt="">
+				{{ site.title }}
+			</a>
+			<button class="navbar-toggler" type="button" data-bs-toggle="collapse" data-bs-target="#navMain" aria-controls="navMain" aria-expanded="false" aria-label="Toggle navigation">
+				<span class="navbar-toggler-icon"></span>
+			</button>
+			<div class="collapse navbar-collapse justify-content-end" id="navMain">
+				<ul class="navbar-nav">
+					{% for item in site.nav %}
+					<li class="nav-item">
+						{% if item.page %}
+						<a class="nav-link{% if page.url == item.page %} active{% endif %}" href="{{ item.page | relative_url }}">
+							{% if item.icon %}<i class="fa-solid {{ item.icon }}"></i>{% endif %} {{ item.title }}
+						</a>
+						{% else %}
+						<a class="nav-link" href="{{ item.url }}" target="_blank" rel="noopener">
+							{% if item.icon %}<i class="fa-solid {{ item.icon }}"></i>{% endif %} {{ item.title }}
+						</a>
+						{% endif %}
+					</li>
+					{% endfor %}
+				</ul>
+			</div>
+		</div>
+	</nav>
+
+	<main class="flex-grow-1" style="margin-top: 4.5rem;">
+		<div class="container-fluid py-4 py-md-5">
+			<div class="row justify-content-center">
+				<div class="col-12 col-lg-10 col-xl-8">
+					<div class="card shadow-lg">
+						<div class="card-body p-4 p-md-5 site-content">
+							{{ content }}
+						</div>
+					</div>
+				</div>
+			</div>
+		</div>
+	</main>
+
+	<footer class="py-3 bg-dark text-light mt-auto">
+		<div class="container-fluid d-flex flex-wrap justify-content-between align-items-center small gap-2 px-3">
+			<span class="d-flex align-items-center gap-2">
+				<a href="https://theta42.com" target="_blank" rel="noopener">
+					<img width="40" src="{{ '/assets/img/theta42.svg' | relative_url }}" alt="theta42">
+				</a>
+				&copy; {{ 'now' | date: '%Y' }} theta42 &middot;
+				<a href="{{ site.github.repository_url }}/blob/master/LICENSE" target="_blank" rel="noopener" class="text-light">MIT License</a>
+			</span>
+			<span class="d-flex align-items-center gap-3">
+				<a href="{{ site.github.repository_url }}" target="_blank" rel="noopener" class="text-light text-decoration-none">
+					<i class="fa-brands fa-github"></i> GitHub
+				</a>
+				<a href="{{ site.github.repository_url }}/blob/master/CHANGELOG.md" target="_blank" rel="noopener" class="text-light text-decoration-none">
+					<i class="fa-solid fa-list"></i> Changelog
+				</a>
+			</span>
+		</div>
+	</footer>
+
+	<script src="https://cdn.jsdelivr.net/npm/bootstrap@5.3.3/dist/js/bootstrap.bundle.min.js"></script>
+</body>
+</html>

docs/agents.md

@@ -0,0 +1,510 @@
+---
+layout: default
+title: Theta Agent & Endpoint Management
+nav_order: 5
+---
+
+# Theta Agent & Endpoint Management
+
+The **Theta Agent** (`theta-agent`) is a unified, 2-way Command & Control (C2) endpoint management daemon written in Go with native cross-platform binaries for **Linux (x86_64, ARM64, ARMv7)**, **Windows (x86_64, ARM64)**, and **macOS (Intel, Apple Silicon M1/M2/M3/M4)**. It connects outbound via a long-lived WebSocket connection to the central **SSO Manager** (`wss://<sso-host>/api/agent/ws`), enabling real-time host telemetry, automated host discovery, and local-first administrative management.
+
+---
+
+## Supported Architectures & Operating Systems
+
+The agent is compiled for 7 target platform binaries with zero external runtime dependencies:
+
+| Operating System | Architecture | Binary Name | Typical Target Devices |
+| :--- | :--- | :--- | :--- |
+| **Linux** | `amd64` (x86_64) | `theta-agent-linux-amd64` | Intel/AMD Servers, Cloud VMs, Proxmox Hypervisors |
+| **Linux** | `arm64` (aarch64) | `theta-agent-linux-arm64` | Raspberry Pi 4/5, Graviton, Ampere Altra |
+| **Linux** | `armv7` (32-bit ARM) | `theta-agent-linux-armv7` | Raspberry Pi 2/3/Zero 2W, ARM IoT Gateways |
+| **Windows** | `amd64` (x86_64) | `theta-agent-windows-amd64.exe` | Windows Server, Windows 10/11 Desktop |
+| **Windows** | `arm64` | `theta-agent-windows-arm64.exe` | Windows on ARM, Surface Pro |
+| **macOS** | `amd64` | `theta-agent-darwin-amd64` | Intel Macs |
+| **macOS** | `arm64` | `theta-agent-darwin-arm64` | Apple Silicon Macs (M1/M2/M3/M4) |
+
+The `install.sh` script automatically detects `uname -s` and `uname -m` to download the exact binary for the host.
+
+---
+
+## Enrollment
+
+An agent is only real if the SSO issued its credential. **Tokens the server did
+not issue are rejected** at the WebSocket handshake.
+
+There are two ways to get a host enrolled, and the first is the normal one.
+
+### Join key — install the agent and the host appears
+
+Hand the machine a **join key** and nothing else. On first connect the SSO
+enrolls the host, issues it its own per-agent token plus the public key it must
+pin, and the agent **writes both into its own `agent.yml`** and blanks the join
+key. From then on it authenticates as itself.
+
+```bash
+curl -fsSL https://<SSO_HOST>/resources/theta-agent/install.sh | sh -s -- \
+  --url "https://<SSO_HOST>" --join-key "tjk_..."
+```
+
+That is the whole procedure — no pre-registering the machine, no copying a
+public key by hand. `setup.sh` mints a key and configures the stack's own host
+this way automatically.
+
+The join key is a *bootstrap* credential, not the host's identity. That
+distinction is what keeps one key convenient without making it a fleet-wide
+skeleton key: every host still ends up individually revocable, and a compromised
+host does not yield a credential that works anywhere else.
+
+| Endpoint | Purpose |
+| :--- | :--- |
+| `GET /api/agent/join-keys` | List keys (prefix + usage only; never the key) |
+| `POST /api/agent/join-keys` | Mint one — returned **once** |
+| `POST /api/agent/join-keys/:id/revoke` | Stop it enrolling new hosts |
+| `DELETE /api/agent/join-keys/:id` | Remove it |
+| `GET /api/agent/join-keys/:id/agents` | Which hosts enrolled through this key |
+
+Revoking a join key does **not** disconnect hosts that already joined; they hold
+their own tokens by then. Revoke the agent itself to cut a specific host off.
+
+**Reuse.** Yes — a join key is not consumed on use. `AgentJoinKey.authenticate`
+only checks `revoked` and `expires_on`; it never invalidates the key itself.
+Every use increments `use_count` and stamps `last_used_on`, but the key keeps
+working until you revoke or delete it (or it expires) — "one key works for as
+many hosts as you like" above is literal, not a figure of speech.
+
+**UI.** The **Install Agent** modal (Directory → Install Agent → Join key tab)
+has a **Manage join keys** table below the mint/select dropdown: label, prefix,
+created date, hosts joined, status, and **Revoke**/**Delete** actions per key.
+Clicking a key's "N hosts" link expands the list of hosts that joined through
+it (name, online status, joined date, last seen).
+
+**Audit.** Yes, both halves are logged as structured `"component":"agent"`
+lines, and the hosts-joined list in the UI above is queryable directly:
+- Minting: `action: "join_key_issued"` records the acting admin (`actor`),
+  `label`, and `keyPrefix`.
+- Each enrollment through that key: `action: "join"` records `agentId`,
+  `agentName`, `remoteAddr`, `joinKeyLabel`, and `joinKeyPrefix`.
+- `GET /api/agent/join-keys/:id/agents` returns the same "which hosts did key
+  X add" answer the UI shows — it matches on the trace `Agent.enroll` leaves in
+  each agent's `description` ("Self-enrolled with join key `<prefix>`") rather
+  than a stored foreign key, since a join key is exchanged for a per-agent
+  token immediately and from then on the agent's own identity is what matters.
+
+### Pre-registering a host
+
+When you want the agent bound to a specific Directory host up front, enroll it
+from **Directory → Install Agent**:
+
+1. Give the agent a name and **bind it to a host resource**. The binding is what
+   links telemetry, status and commands to a Directory entry.
+2. Press **Enroll & issue token**. The SSO mints a 256-bit token, stores only its
+   SHA-256, and shows the raw value **once**.
+3. Copy the generated install command — it already carries the token and the
+   server's public key.
+
+A host that self-enrolls with a join key arrives unbound; bind it afterwards with
+`PUT /api/agent/nodes/:id` or from the Directory.
+
+Or via the API:
+
+```bash
+curl -X POST https://<SSO_HOST>/api/agent/enroll \
+  -H "Authorization: Bearer <admin-api-token>" \
+  -H 'Content-Type: application/json' \
+  -d '{"name": "web01", "resourceId": "<host-resource-uuid>"}'
+```
+
+The response contains `token` (once only) and `publicKey`.
+
+| Endpoint | Purpose |
+| :--- | :--- |
+| `GET /api/agent/nodes` | Every enrolled agent, connected or not, plus the server public key |
+| `POST /api/agent/enroll` | Mint an agent + token |
+| `PUT /api/agent/nodes/:id` | Rename, or bind/unbind the host resource |
+| `POST /api/agent/nodes/:id/rotate` | Issue a new token; the old one stops working immediately |
+| `POST /api/agent/nodes/:id/revoke` | Disable the enrollment |
+| `DELETE /api/agent/nodes/:id` | Remove the enrollment |
+| `POST /api/agent/nodes/:id/command` | Send a command (signed automatically when high-risk) |
+
+Revoke, rotate and delete **drop any live connection immediately** — they do not
+wait for the agent to reconnect. Commands are addressed by agent **id**, never by
+token: a token is a credential and has no business in a URL or a log.
+
+Enrollment, revocation, rotation, every command, and every rejected connection
+are written to the application log as structured `"component":"agent"` records
+with the acting user.
+
+> **Lost the token?** It cannot be recovered — only its hash is stored. Rotate
+> the agent to issue a new one.
+
+---
+
+## Core Functionality
+
+### 1. Host Discovery & Inventory
+Upon establishing a WebSocket connection, the agent immediately pushes a comprehensive discovery payload:
+- **Hostname & Network Interfaces**: Hostname and all non-loopback IPv4 addresses and MACs.
+- **Operating System & Kernel**: Linux distribution, platform, and kernel version.
+- **Hardware Specs**: CPU model, total RAM (GB), and total root disk capacity (GB).
+- **Physical Location**: Location identifier string (e.g. `dc-01-rack-12`) configured in `agent.yml`.
+
+If the agent detects a network IP change, it automatically re-pushes an updated discovery payload to the SSO Manager.
+
+### 2. Real-Time Telemetry Streaming
+Every 30 seconds, the agent streams real-time performance metrics:
+- **CPU Load**: System-wide CPU utilization percentage.
+- **Memory Utilization**: RAM usage percentage and available memory.
+- **Disk Utilization**: Root filesystem usage percentage.
+- **ZFS Storage Health**: Health status of ZFS pools (e.g., `ONLINE`).
+- **NVIDIA GPU Load**: GPU compute utilization percentage (via `nvidia-smi`).
+
+---
+
+## Viewing in the SSO Manager
+
+Agent status and telemetry live on the **Directory** page — there is no separate
+Agents page. For each **host** resource that has a connected theta-agent, the
+Directory shows a status dot in the row:
+
+| Color | Meaning |
+| :--- | :--- |
+| **Green** | Connected, healthy (CPU/RAM/disk within limits). |
+| **Yellow** | Connected but under high load (CPU > 80% or RAM > 80% or disk > 90%). |
+| **Red** | **Enrolled but not connected.** The agent exists and is expected — this is a fault. |
+| **Grey** | No agent enrolled for this host, the enrollment is revoked, or the agent service is unreachable. |
+
+Red and grey used to be the same colour, which made an ordinary directory of
+hosts look like an outage. Because the enrollment now outlives the connection,
+"installed but down" is distinguishable from "never had an agent".
+
+Opening a host's resource modal reveals a **Metrics** tab with the agent's live
+telemetry (CPU/RAM/disk/ZFS/GPU) and discovery info (OS, kernel, IPs, location).
+
+An agent attaches to its host by its **enrollment binding** (`resourceId`), set
+when you enroll it or later via `PUT /api/agent/nodes/:id`. Agents enrolled
+without a binding fall back to matching their reported hostname against the
+resource name — the old behaviour, kept only as a fallback, because it silently
+failed whenever a Directory name differed from the machine's hostname and
+aliased two hosts that happened to share one.
+
+### Agent discovery feeds the Directory
+
+A bound agent's discovery payload is written onto its host resource (`os`,
+`kernel`, `cpu`, `ram_total_gb`, `disk_total_gb`, `ip`), tagged with
+`discovery_sources: ["theta-agent"]` and an `agentId` back-reference. An agent
+runs *on* the host it describes, so it is the most authoritative source the
+directory has. An unbound agent goes through the normal discovery reconciler
+instead, matching like any other source.
+
+---
+
+## Local-First Security & Capability Matrix
+
+To protect hosts against unauthorized control, `theta-agent` enforces a **strict, local-first capability matrix** defined in `/etc/theta42/agent.yml`. Central SSO Manager requests are checked against local configuration before execution; permissions cannot be overridden remotely.
+
+| Capability | Config Key | Risk Level | Description & Impact |
+| :--- | :--- | :--- | :--- |
+| **Telemetry** | `telemetry` | Safe | Streams read-only system metrics (CPU, RAM, Disk, ZFS, GPU). |
+| **Configure LDAP** | `configure_ldap` | Moderate | Writes updated SSSD configuration to `/etc/sssd/sssd.conf` & restarts `sssd`. |
+| **Service Control** | `service_control` | High | Restarts systemd services listed in an explicit allowlist (e.g., `["nginx", "docker", "sssd"]`). |
+| **Reboot** | `reboot` | High | Triggers an immediate system reboot (`systemctl reboot`). |
+| **Arbitrary Bash** | `arbitrary_bash` | Critical | Executes raw bash scripts sent from the SSO Manager as `root` (used for automated GitOps). |
+| **LDAP Tunnel** | `ldap_tunnel` | Moderate | Serves a local LDAP byte-pump socket (`ldap_socket`, default `/run/theta/ldap.sock`) for SSSD/PAM. The agent never parses LDAP — it forwards raw bytes to the SSO, which relays them into its own OpenLDAP. |
+| **Secrets** | `secrets` | Moderate | Renders OpenBao secrets to local files from templates (see [Secrets Engine](#secrets-engine---rendering-openbao-secrets-to-local-files) below). |
+| **IAM** | `iam` | Critical | Applies SSO-pushed node identity config: sudo rules, SSH `AuthorizedKeysCommand` keys, `/etc/security/access.conf`, and revocation (`sss_cache -E` + session kill). Every push is Ed25519-signed. |
+
+---
+
+## High-Risk Command Verification (Protocol v1.2.0)
+
+High-risk management commands (`reboot`, `service_restart`, `configure_ldap`, `arbitrary_bash`, `update_binary`) are cryptographically verified using **Ed25519 signatures**:
+1. The SSO Manager canonicalizes the command payload (sorted keys, no whitespace,
+   no HTML escaping, `signature` omitted).
+2. The payload is signed with the SSO Manager's Ed25519 private key.
+3. The Base64 signature is appended to the message payload.
+4. The agent verifies the signature against the configured `public_key` in `/etc/theta42/agent.yml` before executing the action.
+
+**The signing key is persistent.** It lives in OpenBao at
+`secret/agent/signing-key` and survives restarts, so the `public_key` you pin in
+`agent.yml` keeps matching. (It used to be generated in memory at boot and
+changed on every restart, which made pinning impossible.) If the SSO cannot load
+or store a key it **refuses** to send high-risk commands rather than signing with
+one no agent has seen — `GET /api/agent/nodes` reports this as
+`signingAvailable: false`.
+
+This requires the `sso-broker` OpenBao policy to grant `secret/agent/*`. Re-run
+`./setup.sh` from theta-suite if you are upgrading.
+
+**Verification is fail-closed on the agent.** An agent with no `public_key`
+configured rejects every high-risk command. Earlier versions logged "skipping
+signature verification" and executed them, so an agent installed without a key
+would run `reboot`, `configure_ldap` and `arbitrary_bash` unverified.
+
+---
+
+## Secrets Engine — rendering OpenBao secrets to local files
+
+The agent can render OpenBao secrets to local files that any process on the
+host — a bash script, a systemd unit, a Node app, whatever — reads like an
+ordinary env file. The agent never holds a Vault token: it asks the SSO for the
+values over its existing WSS channel, and the SSO fetches them from OpenBao
+using its own access, scoped so the agent can only ever read its own node's
+secrets.
+
+**Node scope.** Every path an agent can request must start with
+`secret/data/nodes/<this-agent's-id>/`. The SSO enforces this server-side
+(`POST /api/v1/agent/secrets`); a request for any other node's path is
+rejected:
+
+```
+$ curl -sk https://sso.example.com/api/v1/agent/secrets \
+    -H "Authorization: Bearer <agent-token>" -H 'Content-Type: application/json' \
+    -d '{"paths":["secret/data/nodes/some-other-node-id/db"]}'
+{"status":"error","message":"path outside node scope: secret/data/nodes/some-other-node-id/db"}
+```
+
+A compromised agent can therefore never reach another host's secrets, or
+anything outside `secret/data/nodes/*`.
+
+### Walkthrough: a 3rd-party app reads a secret the agent rendered
+
+This walks through the whole path end to end, on a stack freshly brought up
+from theta-suite's own `docs/fixtures.md` demo data — the same steps work on
+any theta-suite install.
+
+**1. Enroll the host.** Directory → Install Agent → mint a join key, run the
+install command on the target host as root.
+
+<a href="images/agent-install-join-key.png" target="_blank"><img src="images/agent-install-join-key.png" alt="Install Theta Agent modal with a freshly minted join key and install command" width="80%"></a>
+
+On first connect the agent exchanges the join key for its own token + the
+SSO's public key and writes both back into `/etc/theta42/agent.yml`. Note the
+agent's id from `GET /api/agent/nodes` (or the Directory URL) — you need it for
+the next step.
+
+**2. Turn on the `secrets` capability and point it at a template.** Add to the
+host's `/etc/theta42/agent.yml`:
+
+```yaml
+secrets:
+  - template: /etc/theta/templates/db.env.tpl
+    target: /etc/theta/rendered/db.env
+    reload: ""          # optional: e.g. "systemctl reload myapp"
+
+capabilities:
+  secrets: true
+```
+
+And the template itself, `/etc/theta/templates/db.env.tpl` — placeholders are
+`{{ bao "secret/data/nodes/<agent-id>/<name>#<key>" }}`:
+
+```
+DB_USER="{{ bao "secret/data/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/db#username" }}"
+DB_PASS="{{ bao "secret/data/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/db#password" }}"
+```
+
+Restart the agent to pick up the config change.
+
+**3. Seed the secret.** From `theta-suite/` (theta-env), as the operator:
+
+```
+./setup.sh --seed-node-secret f9a30ab0-7d8a-4b77-a4c4-6a6383d084db db \
+  username=demoapp password=CorrectHorseBattery42
+```
+
+This writes to `secret/nodes/<agent-id>/db` in OpenBao (the CLI path — the HTTP
+API the agent uses sees it as `secret/data/nodes/<agent-id>/db`, matched by the
+node-scope check above). It's idempotent: it skips silently if that path is
+already seeded.
+
+**4. Trigger the render.** The Directory UI doesn't have a button for this yet
+— push it the same way any admin command goes out, `POST
+/api/agent/nodes/:id/command`. It's in the high-risk list, so the SSO signs it
+automatically:
+
+```
+curl -X POST https://sso.example.com/api/agent/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/command \
+  -H "auth-token: <admin session token>" -H 'Content-Type: application/json' \
+  -d '{"command": "render_secrets", "payload": {}}'
+```
+
+The agent logs `Received command: render_secrets` / `Rendering secret
+templates...` and atomically writes the target file at mode `0600`:
+
+```
+$ cat /etc/theta/rendered/db.env
+DB_USER="demoapp"
+DB_PASS="CorrectHorseBattery42"
+```
+
+Back in the Directory, the host's Metrics tab shows **Secrets** lit up green
+among the reported capabilities:
+
+<a href="images/agent-capabilities-metrics.png" target="_blank"><img src="images/agent-capabilities-metrics.png" alt="Directory Metrics tab showing live telemetry and the agent's reported capability badges, with Telemetry and Secrets lit green" width="80%"></a>
+
+**5. Read it from a bash app on the same host.** The rendered file is just an
+env file — no agent involvement needed to consume it:
+
+```sh
+#!/bin/sh
+. /etc/theta/rendered/db.env
+echo "DB_USER=$DB_USER"
+echo "DB_PASS=$DB_PASS"
+```
+
+**6. Read it from a Node app on the same host:**
+
+```js
+const fs = require('fs');
+const env = fs.readFileSync('/etc/theta/rendered/db.env', 'utf8');
+const db = {};
+for (const line of env.split('\n')) {
+  const m = /^(\w+)="(.*)"$/.exec(line.trim());
+  if (m) db[m[1]] = m[2];
+}
+console.log('DB_USER=' + db.DB_USER);
+console.log('DB_PASS=' + db.DB_PASS);
+```
+
+Both print the same values the template resolved — `demoapp` /
+`CorrectHorseBattery42` in this walkthrough. `theta-agent/demo/` in the
+theta-agent repo has these two scripts ready to run.
+
+### Alternative: calling the API directly
+
+Rendering to a file is the normal path — it works for any app regardless of
+language, and the secret never touches an HTTP client the app itself controls.
+But an app can also fetch its node's secrets directly, bypassing the template
+engine entirely (useful for debugging, or a process that wants to hold the
+value only in memory). This uses the **agent's own bearer token**, not an admin
+token — the same node-scope enforcement applies:
+
+```sh
+curl -sk https://sso.example.com/api/v1/agent/secrets \
+  -H "Authorization: Bearer <agent-token>" -H 'Content-Type: application/json' \
+  -d '{"paths":["secret/data/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/db"]}'
+```
+
+```js
+const token = process.env.THETA_AGENT_TOKEN; // from /etc/theta42/agent.yml
+fetch('https://sso.example.com/api/v1/agent/secrets', {
+  method: 'POST',
+  headers: { Authorization: 'Bearer ' + token, 'Content-Type': 'application/json' },
+  body: JSON.stringify({ paths: ['secret/data/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/db'] })
+}).then(r => r.json()).then(d => console.log(d.secrets));
+```
+
+Both return:
+
+```json
+{
+  "status": "ok",
+  "secrets": {
+    "secret/data/nodes/f9a30ab0-7d8a-4b77-a4c4-6a6383d084db/db": {
+      "username": "demoapp",
+      "password": "CorrectHorseBattery42"
+    }
+  }
+}
+```
+
+---
+
+## Installation & Deployment
+
+### Quick One-Liner Install
+Run the following command as `root` on the target Linux host:
+
+```bash
+curl -fsSL https://<SSO_HOST>/resources/theta-agent/install.sh | sh -s -- \
+  --url "https://<SSO_HOST>" --token "<ISSUED_TOKEN>" --public-key "<BASE64_PUBLIC_KEY>"
+```
+
+Both values come from enrollment. The **Install Agent** modal builds this line
+for you with them already filled in. Omitting `--public-key` leaves the agent
+able to report telemetry but unable to accept any high-risk command.
+
+### Custom Config Wizard
+You can generate a Base64-encoded custom configuration using the **Install Agent** button on the **Directory Management** page in the SSO Manager UI:
+
+```bash
+curl -fsSL https://<SSO_HOST>/resources/theta-agent/install.sh | sh -s -- "<BASE64_ENCODED_CONFIG>"
+```
+
+---
+
+## Configuration File Example (`/etc/theta42/agent.yml`)
+
+```yaml
+# /etc/theta42/agent.yml
+server_url: "wss://sso.example.com"
+# Issued by the SSO. Left empty when installing with a join key -- the agent
+# fills it in itself once the server enrolls it.
+auth_token: "c8181ce0e55bf7302b11d719a7ae39adcd7604de461e6e363f8bb4fadf126acb"
+# Bootstrap credential. Used only while auth_token is empty, and blanked by the
+# agent once it has its own token.
+join_key: ""
+location: "dc-01-rack-12"
+# Base64 of the RAW 32-byte Ed25519 public key -- exactly the `publicKey` value
+# from enrollment or GET /api/agent/nodes. Not a PEM body: a base64-decoded
+# SPKI blob is 44 bytes, the agent requires 32, and it will refuse every signed
+# command if this is wrong.
+public_key: "D0cJB3iuStTzhXlu7tFDh/eEXFxRZwkuwQJJhFSqwlQ="
+
+capabilities:
+  telemetry: true
+  configure_ldap: true
+  reboot: false
+  service_control: ["nginx", "docker", "sssd"]
+  arbitrary_bash: false
+```
+
+---
+
+## Troubleshooting: agent is rejected (`close 4001`)
+
+If the agent logs that the server rejected its token, the enrollment — not the
+network — is the problem. The SSO accepts the WebSocket upgrade and then closes
+with an application code:
+
+| Code | Meaning | Fix |
+| :--- | :--- | :--- |
+| `4001` | Token unknown, or never issued by this server | Enroll the host and put the issued token in `agent.yml` |
+| `4002` | Superseded — another connection authenticated as this agent | Normal; two copies of the agent are running |
+| `4003` | Enrollment revoked or deleted | Re-enroll |
+| `4004` | Token rotated; `agent.yml` has the old value | Copy the new token |
+
+The agent backs off for 5 minutes on `4001`/`4003`/`4004` rather than retrying
+every 5 seconds — a credential that is wrong will not fix itself, and hammering
+the SSO only floods its audit log.
+
+An agent installed before protocol v1.2.0 carries a token generated in the
+browser that the server never recorded, so it will be rejected with `4001` until
+re-enrolled. The quickest fix is to put a **join key** in its `agent.yml` as
+`join_key` and blank `auth_token` — it will re-enroll itself on the next
+reconnect.
+
+---
+
+## Troubleshooting: agent can't connect (`dial tcp ... i/o timeout`)
+
+If the agent host logs `Dial error: dial tcp <ip>:443: i/o timeout` while
+connecting to `wss://<sso-host>/api/agent/ws`, the WebSocket path is usually
+fine — this is a **network/NAT** problem, not an agent or SSO bug. A host behind
+the same NAT that owns the SSO often cannot reach its own **public IP** (no
+hairpin/loopback NAT on many home routers), so the TCP dial times out even
+though the same address works from outside.
+
+Fix options:
+1. Point `agent.yml` `server_url` at an address the host can reach directly —
+   e.g. the SSO host's LAN IP (`http://<lan-ip>` or `http://<lan-ip>:3001` for a
+   no-TLS direct path).
+2. Enable **NAT reflection / hairpin NAT** on the router so LAN hosts can reach
+   their own public IP:443.
+3. Add a local route/firewall rule on the agent host for its public IP.
+
+> Note: on a deployment where the theta42 proxy fronts `sso.suite.example`, make
+> sure the proxy has a **persistent Host record** for the real SSO domain — not
+> just the `localtest.me` placeholder — so routing survives a proxy restart
+> (an in-memory lookup cache can mask a missing Redis record for up to ~1h).

docs/assets/css/style.css

@@ -0,0 +1,116 @@
+/* theta42 docs site — shares the in-app dark navbar/footer + card look
+   (Bootstrap 5 + Font Awesome, same as the running apps) rather than a
+   generic Jekyll theme. */
+
+body {
+	background-color: #f4f5f6;
+}
+
+.navbar-brand img {
+	filter: drop-shadow(0 0 2px rgba(0, 0, 0, .4));
+}
+
+.navbar-nav .nav-link.active {
+	color: #fff;
+	font-weight: 600;
+}
+
+/* Markdown content typography, scoped to the card body so it doesn't leak
+   into the nav/footer. */
+.site-content h1:first-child {
+	margin-top: 0;
+}
+
+.site-content h1,
+.site-content h2,
+.site-content h3 {
+	font-weight: 700;
+}
+
+.site-content h2 {
+	margin-top: 2.5rem;
+	padding-bottom: .4rem;
+	border-bottom: 1px solid #e9ecef;
+}
+
+.site-content h3 {
+	margin-top: 1.75rem;
+}
+
+.site-content a {
+	color: #a3671f;
+	text-decoration-color: rgba(163, 103, 31, .35);
+}
+
+.site-content a:hover {
+	color: #8a5a16;
+}
+
+.site-content pre {
+	background-color: #212529;
+	color: #f8f9fa;
+	padding: 1rem 1.25rem;
+	border-radius: .375rem;
+	overflow-x: auto;
+}
+
+.site-content code {
+	color: #a3671f;
+	background-color: #f4f0e8;
+	padding: .15em .4em;
+	border-radius: .25rem;
+	font-size: .875em;
+}
+
+.site-content pre code {
+	color: inherit;
+	background: none;
+	padding: 0;
+}
+
+.site-content table {
+	display: block;
+	overflow-x: auto;
+	width: 100%;
+	border-collapse: collapse;
+	margin: 1.25rem 0;
+}
+
+.site-content table th,
+.site-content table td {
+	border: 1px solid #dee2e6;
+	padding: .5rem .75rem;
+	text-align: left;
+}
+
+.site-content table th {
+	background-color: #f8f9fa;
+}
+
+.site-content blockquote {
+	border-left: 4px solid #C59341;
+	padding: .5rem 1rem;
+	margin: 1.25rem 0;
+	background-color: #f8f6f1;
+	color: #495057;
+}
+
+.site-content img {
+	max-width: 100%;
+	height: auto;
+}
+
+/* Screenshot grids in the markdown use width="49%" inline attrs for a
+   two-up desktop layout -- stack them on narrow screens instead of
+   squeezing to illegibility. */
+@media (max-width: 576px) {
+	.site-content img[width] {
+		width: 100% !important;
+		margin-bottom: .75rem;
+	}
+}
+
+.site-content hr {
+	margin: 2rem 0;
+	border-top: 1px solid #e9ecef;
+}

docs/assets/img/theta42.svg

@@ -0,0 +1,51 @@
+<svg xmlns="http://www.w3.org/2000/svg" viewBox="0 0 400 400" width="100%" height="100%">
+  <defs>
+    <linearGradient id="gold-grad" x1="0%" y1="0%" x2="100%" y2="100%">
+      <stop offset="0%" stop-color="#C59341" />
+      <stop offset="20%" stop-color="#E4B869" />
+      <stop offset="40%" stop-color="#FBF0B9" />
+      <stop offset="60%" stop-color="#DFB260" />
+      <stop offset="80%" stop-color="#BC8837" />
+      <stop offset="100%" stop-color="#A36F28" />
+    </linearGradient>
+
+    <linearGradient id="text-grad" x1="0%" y1="100%" x2="100%" y2="0%">
+      <stop offset="0%" stop-color="#FFFFFF" />
+      <stop offset="40%" stop-color="#F5E3B5" />
+      <stop offset="70%" stop-color="#D4A343" />
+      <stop offset="100%" stop-color="#8A5A16" />
+    </linearGradient>
+
+    <filter id="drop-shadow" x="-20%" y="-20%" width="140%" height="140%">
+      <feDropShadow dx="0" dy="8" stdDeviation="6" flood-color="#000000" flood-opacity="0.4"/>
+    </filter>
+  </defs>
+
+  <g filter="url(#drop-shadow)">
+    <g fill="url(#gold-grad)">
+      <path d="M 200,40 
+               C 290,40 350,110 350,200 
+               C 350,290 290,360 200,360 
+               C 110,360 50,290 50,200 
+               C 50,110 110,40 200,40 Z 
+               M 200,75 
+               C 130,75 88,130 88,200 
+               C 88,270 130,325 200,325 
+               C 270,325 312,270 312,200 
+               C 312,130 270,75 200,75 Z" 
+            fill-rule="evenodd" />
+      
+      <path d="M 88,190 L 140,190 C 140,190 142,210 140,210 L 88,210 Z" />
+      
+      <path d="M 260,190 L 312,190 C 312,190 310,210 260,210 Z" />
+    </g>
+
+    <text x="200" y="222" 
+          font-family="system-ui, -apple-system, 'Segoe UI', Roboto, Helvetica, Arial, sans-serif" 
+          font-size="78" 
+          font-weight="900" 
+          fill="url(#text-grad)" 
+          text-anchor="middle"
+          letter-spacing="-2">42</text>
+  </g>
+</svg>

docs/concepts-accounts.md

@@ -0,0 +1,125 @@
+---
+layout: default
+title: Accounts, Groups & Managers
+description: A plain-language guide to users, service accounts, personal groups, and managers in SSO Manager.
+---
+
+# Accounts, Groups & Managers
+
+This page explains the concepts behind the Users and Groups pages in plain
+language. If you want the technical schema/attribute-level detail instead,
+see the [LDAP reference](ldap.html).
+
+## What's an account?
+
+Every person (or app) that can sign in through this SSO Manager has an
+**account** — a username, a display name, maybe an email address, and a
+password (or, for service accounts, no password at all — see below).
+Accounts live in the directory this app manages, and any other app you've
+connected (Gitea, Home Assistant, your Wi-Fi, whatever) checks against these
+same accounts instead of keeping its own separate list of users and
+passwords.
+
+## Two kinds of account: people and service accounts
+
+Most accounts belong to an actual person — check **Users → People** to see
+them. But sometimes you need an account for something that *isn't* a
+person: a media server, a backup script, a bind account another app uses to
+look people up. These are **service accounts**, listed separately under
+**Users → Service Accounts**, and they're different from a person's account
+in two ways that matter:
+
+- **No email required.** A service account doesn't need a mailbox, so the
+  form doesn't ask for one.
+- **A password is optional.** If you leave it blank, nobody can log in as
+  that account — which is exactly what you want for something that only
+  ever gets used programmatically (a script authenticating with an API
+  token, or another app binding with a fixed, separately-configured
+  password you set yourself). Only give it a password if the account
+  genuinely needs to log in or bind somewhere as itself.
+
+Aside from those two differences, a service account is a completely normal
+account under the hood — it can belong to groups, have a manager, and so
+on, just like anyone else's.
+
+## Groups: who can do what
+
+A **group** is just a named list of accounts, used to control access. This
+app has a handful of built-in groups that grant admin powers (e.g. only
+people in the `app_sso_admin` group can see the Users/Groups/Directory/Overview
+pages at all), but you can also make your own groups for any app you
+connect — say, a group listing everyone who should be allowed into your
+photo server. Once a group exists, add or remove members from the
+**Groups** page, and point the other app's "who's allowed in" setting at
+that group's name.
+
+### Groups inside groups
+
+A group can contain another group, not just people — the *Nested* tab on any
+group card. Everyone in the inner group counts as a member of the outer one,
+however many levels deep it goes.
+
+This is mostly a way to stop repeating yourself. Make one `developers` group,
+nest it into the handful of things developers should reach, and adding a new
+developer to that one group grants all of them at once — instead of adding them
+to each individually and slowly drifting out of sync. The app already does this
+for itself: super admins are nested into every resource's admin group, and each
+admin group into its access group, so "can administer it" always implies "can
+use it".
+
+Two things it won't let you do: put a group inside itself (directly or round a
+longer loop), and empty a group completely — every group must keep at least one
+member.
+
+A note if you also manage the directory by hand: a group's member list shows
+what is *directly* listed on it. Someone who gets in through a nested group is
+a real member but won't appear there — the **Nested** tab shows what is nested,
+and the API's `effective` view lists everyone who actually gets in.
+
+## Every account's personal group
+
+Separately from the groups above, every single account — person or
+service account — automatically gets its own small, personal group when
+it's created, named after the account itself. Most of the time you'll
+never think about this; it exists so that, on a Linux system connected to
+this directory, each account "owns" its own files by default the same way
+a normal Unix user account would.
+
+Occasionally you'll want to share that ownership with someone else — for
+example, letting a second account also have write access to files a
+service account owns. That's what the **"Members of `<uid>`'s group"**
+section on a profile page is for: add another account there, and the
+underlying Linux permissions treat them as if they belong to that same
+personal group too.
+
+## What's a "manager"?
+
+Every account has one or more **managers** — the people allowed to edit
+that account's profile (phone number, SSH key, home directory, and so on)
+without needing full admin rights. By default, whoever created an account
+(the admin who added it, or whoever sent the invite) becomes its first
+manager, but you can add or remove managers later from the account's Edit
+form.
+
+This is useful for service accounts especially: if a service account
+belongs to a particular project or person, make them its manager so they
+can maintain it — rotate its SSH key, adjust its description — without
+needing to be a full SSO administrator.
+
+## Inviting someone vs. adding them yourself
+
+From the Users page you can either fill in someone's details yourself
+("Add new user"), or send them an **invite** — an email (or a link you copy
+and send however you like) that lets them pick their own username and
+password. Either way, the resulting account is identical; invites are just
+a convenience so you don't have to know someone's preferred username or
+handle their password directly.
+
+## Want more detail?
+
+This page deliberately leaves out LDAP schema names, attribute types, and
+protocol-level detail. If you're connecting a third-party app directly to
+the LDAP directory, or you just want to know exactly what's stored where,
+see the [LDAP reference](ldap.html).
+
+[← Back to Home](index.html)

docs/concepts-api-tokens.md

@@ -0,0 +1,59 @@
+---
+layout: default
+title: API Tokens
+description: A plain-language guide to personal access tokens in SSO Manager.
+---
+
+# API Tokens
+
+This page explains what an API token is and when you'd want one. For the
+full list of API endpoints a token can call, see the
+[API reference](https://github.com/theta42/sso-manager-node/blob/master/API.md).
+
+## What's an API token, in plain terms?
+
+Normally, you interact with this app by logging in through a web browser.
+An **API token** (also called a personal access token, or PAT) is an
+alternative way in — a long, random string that a script, a scheduled job,
+or another program can use instead of a username and password, to act on
+your behalf without a human typing a login in each time.
+
+If you've ever set up a script to talk to GitHub, GitLab, or a similar
+service using a "token" instead of your real password, this is the same
+idea.
+
+## When would you actually need one?
+
+Most people never need to create one of these — you'll only want a token
+if you're automating something, for example:
+
+- A script that syncs users or groups from somewhere else into this SSO
+  Manager on a schedule.
+- A backup or monitoring job that checks this app's health via its API.
+- A CI/CD pipeline that needs to register or update an OAuth client
+  automatically.
+
+If you're not doing any of that, you don't need an API token — just log in
+normally through the web UI.
+
+## How it works
+
+Create a token from your Profile page, give it a name so you remember what
+it's for later, and optionally an expiry. You'll be shown the token's
+value **exactly once** — copy it somewhere safe immediately, because it
+can't be viewed again afterward (only revoked or rotated). Whatever script
+or tool you're using it with sends it along with each request, the same
+way a browser sends your login session.
+
+A token acts **as you**, with **your** permissions — if you're not an
+admin, a token you create can't do admin-only things either. If you ever
+suspect a token has leaked (ended up somewhere it shouldn't have, like a
+public script or log file), revoke it immediately from your Profile page;
+it stops working right away.
+
+## Want more detail?
+
+This page doesn't attempt to list every API endpoint or show request/
+response examples — for that, see the full [API reference](https://github.com/theta42/sso-manager-node/blob/master/API.md).
+
+[← Back to Home](index.html)

docs/concepts-oauth-apps.md

@@ -0,0 +1,79 @@
+---
+layout: default
+title: Connecting Apps (Single Sign-On)
+description: A plain-language guide to OAuth/OIDC clients and single sign-on in SSO Manager.
+---
+
+# Connecting Apps (Single Sign-On)
+
+This page explains, in plain language, what happens when you "connect" an
+app to your SSO Manager so people can log into it with their existing
+account. For the technical endpoint/token detail, see the
+[OAuth reference](oauth.html).
+
+## What does "single sign-on" actually mean?
+
+Instead of every app you run having its own separate list of usernames and
+passwords, they all check with this SSO Manager instead. You log in once,
+here, and any connected app trusts that login — no separate password to
+remember or manage for each one. If you ever need to lock someone out
+everywhere at once, you do it in one place (deactivate their account here)
+instead of hunting down every app individually.
+
+The technology behind this is called **OAuth 2.0** and **OpenID Connect
+(OIDC)** — you'll see both names used, often together, referring to the
+same thing. You don't need to understand the protocol to use this page;
+what matters practically is the handful of concepts below.
+
+## What's a "client"?
+
+Every app you connect is registered here as a **client** — a single entry
+in the Directory representing that one app. Registering a client
+gives you a **Client ID** and **Client Secret**: think of these like a
+username and password, but for the *app itself* rather than for a person.
+You paste them into the other app's own "Single Sign-On" or "OIDC" setup
+screen, along with the discovery URL shown at the top of this page, and
+that app is now able to ask this SSO Manager to authenticate people on its
+behalf.
+
+**Treat the Client Secret like a password** — anyone who has it can
+impersonate that app when talking to your SSO Manager. If you ever suspect
+it's leaked, rotate it from the client's card.
+
+## What are "scopes"?
+
+**Scopes** control what information a connected app is allowed to ask for
+about the person logging in — their username, email, group memberships,
+and so on. Most apps tell you exactly which scopes they need in their own
+setup instructions; when in doubt, the default set (`openid`, `profile`,
+`email`, `groups`) covers what nearly every app expects.
+
+## "Restrict to Groups"
+
+By default, *any* account with an SSO Manager login can sign into a
+connected app. If that's not what you want — say, a home automation
+dashboard that only certain family members should reach — set **Restrict
+to Groups** on that client to one of your [groups](concepts-accounts.html).
+Only members of that group will be allowed to log into that particular
+app; everyone else gets turned away at the login step, even though their
+SSO Manager account still works everywhere else.
+
+## Redirect URIs
+
+A **Redirect URI** is the exact web address the connected app wants people
+sent back to once they've logged in here — it's a security measure so an
+attacker can't trick the login flow into redirecting somewhere else. The
+app's own setup instructions will tell you this value; copy it in exactly
+as given. If the app is reachable via more than one hostname (for example,
+because it sits behind [theta42/proxy](https://theta42.github.io/proxy/)),
+this field supports wildcard patterns — see the inline help under the
+field itself for the exact syntax.
+
+## Want more detail?
+
+This page intentionally skips the protocol-level detail (exact endpoint
+URLs, token formats, claim names). If you're troubleshooting a connection
+or building something against the API directly, see the
+[OAuth reference](oauth.html).
+
+[← Back to Home](index.html)

docs/configuration.md

@@ -1,6 +1,7 @@
 ---
 layout: default
 title: Configuration
+description: SSO Manager's config layers — conf/base.js defaults, secrets.js overrides, and app_* environment variables.
 ---
 
 # Configuration
@@ -15,13 +16,27 @@ deep-merges, in order (later wins):
    `localhost`, `SSO Manager`).
 2. `conf/<NODE_ENV>.js` — optional, environment-specific.
 3. `conf/secrets.js` — gitignored; secrets + per-deployment values.
-4. **`app_*` environment variables** — the highest-precedence layer.
+4. **`app_*` environment variables** — the highest-precedence layer among these
+   four.
 
 Any env var whose name starts with `app_` overrides the merged config. The rest
 of the name splits on **double-underscore** (`__`) into a nested path. Values are
 `JSON.parse`-coerced when possible (numbers, booleans, null, JSON) and kept as
 raw strings otherwise.
 
+### A fifth, higher-precedence layer: OpenBao + the Configuration UI
+
+In a theta-suite deployment, `@simpleworkjs/bao-conf`'s `init()` deep-merges
+`secret/sso-manager/conf` (from OpenBao) over the four layers above at boot —
+this is the layer `setup.sh`/theta-suite actually manages, and it wins over
+everything else here. On top of that, the admin **Configuration** page in the
+UI writes straight to `secret/sso-manager/conf` (via `routes/api_conf.js`)
+and applies the change to the live `conf` object immediately
+(`applyToLiveConf`) — no restart, and it bypasses `conf/secrets.js` entirely.
+If a value isn't behaving the way `conf/secrets.js` says it should, check the
+Configuration UI / OpenBao before assuming a file edit didn't take — it's
+almost certainly OpenBao (or a live UI edit) winning the merge.
+
 ## Examples
 
 | Env var | Sets | Type |
@@ -31,6 +46,8 @@ raw strings otherwise.
 | `app_ldap__userBase=ou=people,dc=…` | `conf.ldap.userBase` | string |
 | `app_ldap__uidGidMin=1500` | `conf.ldap.uidGidMin` | number (new-user id floor) |
 | `app_ldap__uidGidReservedFloor=9000` | `conf.ldap.uidGidReservedFloor` | number (ids at/above this are ignored when allocating) |
+| `app_ldap__ldapsHost=ldap.internal.example.com` | `conf.ldap.ldapsHost` | string (hostname shown on `/integrations` for LDAPS binds; empty = derive from `oauth.issuer`) |
+| `app_ldap__ldapsPort=636` | `conf.ldap.ldapsPort` | number (port shown on `/integrations`) |
 | `app_oauth__jwtSecret=...` | `conf.oauth.jwtSecret` | string |
 | `app_oauth__issuer=https://sso.example.com` | `conf.oauth.issuer` | string |
 | `app_oauth__token_lifetime__access_token=3600` | `conf.oauth.token_lifetime.access_token` | number |

docs/deployment.md

@@ -1,6 +1,7 @@
 ---
 layout: default
 title: Deployment
+description: Deploying SSO Manager — the all-in-one Docker image, bare-metal install, config layers, and backups.
 ---
 
 # Deployment Guide

docs/directory.md

@@ -0,0 +1,185 @@
+---
+layout: default
+title: Directory Management
+description: Managing your Home-Lab infrastructure, services, and LDAP access relationships via the SSO Directory API.
+---
+
+# Directory Management
+
+The SSO Manager ships with a built-in **Directory & Inventory Management** feature. Instead of just managing bare LDAP groups for your homelab, the Directory allows you to map out your infrastructure graph and assign rich metadata to your services.
+
+## Architecture
+
+The Directory models your homelab infrastructure using a parent-child graph (e.g. `Site -> Host -> Service`).
+
+There are three primary **Kinds** of resources you can define:
+- **Site**: A physical location, datacenter, or root node (e.g., `us-east`). Sites do not require parents.
+- **Host**: A physical machine, Proxmox node, virtual machine, or LXC container. A Host **must** have a parent Site or another Host.
+- **Service (App)**: An application, web service. A Service **must** have a parent Host or another Service.
+- **OAuth Integration**: An OAuth 2.0 / OpenID Connect client application. An OAuth integration **must** have a parent Service.
+
+By defining this hierarchy, the SSO Manager builds a queryable graph of your infrastructure.
+
+## Automatic LDAP Group Creation
+
+When you create a new **Host** or **Service** in the Directory via the web UI (or API), the SSO Manager will automatically provision two LDAP groups in your directory to govern access to that resource:
+
+1. `<slug>_access` (Member level access)
+2. `<slug>_admin` (Owner level access)
+
+For example, if you create a Service named "Emby" with the slug `app_emby`, the system will create the LDAP groups `app_emby_access` and `app_emby_admin`. You can then assign users to these groups, and they will immediately see the service populate on their "My Services" dashboard.
+
+## Resource Metadata
+
+Resources carry a flexible `metadata` JSON object that can store essential context for your applications. The UI natively supports the following metadata fields:
+
+### Common Metadata
+- **Sub Type**: Free-form text to categorize the resource (e.g., `proxmox_node`, `linux`, `lxc`, `web`).
+- **IP Address**: The internal IP address of the resource.
+- **MAC Address**: The hardware address of the primary interface.
+- **Host / URI Address**: The FQDN or URL of the resource (e.g., `https://emby.home.arpa`).
+- **Production Environment**: A boolean toggle indicating if the resource is in production.
+
+### Host Metadata
+- **VMID**: The hypervisor VM or Container ID (e.g. `101`).
+- **OS**: The operating system name (e.g. `Ubuntu 22.04.3 LTS`).
+- **Kernel**: The kernel version string (e.g. `5.15.0-100-generic`).
+
+### Service Metadata
+- **Internal Port**: The local port the service binds to (e.g. `8080`).
+- **External Port**: The reverse-proxy or external port (defaults to Internal Port if left blank).
+- **Public (No Auth)**: Indicates if the service is exposed publicly without authentication.
+- **External Reachable**: Indicates if the service is accessible outside the VPN/local network.
+- **Git Repo**: The source code repository for the service (e.g. `https://github.com/...`).
+- **Install Path**: The filesystem path where the service is installed (e.g. `/opt/app`).
+- **Systemd Service**: The systemd unit name for the service (e.g. `app.service`).
+
+### Who sees which metadata
+
+Metadata keys are declared in `@simpleworkjs/directory-schema` with an `admin` flag, and every API response is passed through its projection. There are three tiers:
+
+- **Public** — returned to any authenticated caller, including machine (`ServiceToken`) callers: `ip`, `address`, `sshPort`, `fqdn`, `dnsNames`, `port`, `externalPort`, `portMappings`, `isExternalReachable`, `os`, `gitRepo`, `subType`, `icon`, `tagline`, `isPublic`, `isProduction`, `requestable`, `isCurrentSite`.
+- **Admin-only** — only for members of `app_sso_directory_admin` / `app_sso_admin`: `vmid`, `macAddress`, `installPath`, `systemdService`, and the OAuth config keys (`redirect_uris`, `scopes`, `allowed_groups`, `token_lifetime`).
+- **Never returned** — `client_secret_hash`, plus any key matching `/secret|password|privatekey/i`. Stripped on every path, admins included.
+
+Note that machine tokens are deliberately *not* admins, so anything a machine consumer needs (the firewall generator reads `port` / `externalPort` / `isExternalReachable`) has to be in the public tier. A metadata key that isn't declared at all is treated as admin-only and will silently vanish for normal users — if you add a field to the admin form, declare it in the schema package too.
+
+## Catalog & access requests
+
+The site root (`/`) is the end-user catalog — the only ungated page in the nav. It shows:
+
+- **My Access** — everything the signed-in user can reach (`GET /api/discovery/me`), each card carrying a **how to reach it** block: the URL for a service, or the SSH invocation for a host. When `directory.jumpHost` is set in the config, host cards render the jump-host form `ssh <uid>_-_<slug>@<jumpHost>`; otherwise they fall back to a direct `ssh <uid>@<ip>`.
+- **Discover More** — everything else in the directory, with a **Request access** button.
+- **My Requests** / **Awaiting My Approval** — pending requests, and the approve/deny queue for anyone who owns a requested resource.
+
+A request is a proposal to join an LDAP group. It targets the resource's `member`-level group (the `_access` one, never `_admin`), and approving it performs the LDAP group add — so LDAP stays the single access-control truth and the table is just the audit trail. Approvals are idempotent: approving for someone already in the group succeeds rather than erroring.
+
+Requests are decided by the resource's `owner`, or by any directory admin. Mark a resource `metadata.requestable = false` to keep it out of self-service.
+
+## Navigating the UI
+
+The Directory Management interface nests your resources as a tree, making it easy
+to comprehend your network topography at a glance. You can filter, search, and
+sort your entire infrastructure inventory. Click the green `+` icon next to any
+resource to add a child resource beneath it.
+
+**Collapsing the tree.** Any resource with children carries a caret; click it to
+fold that subtree away. The toolbar's double-chevron buttons expand or collapse
+everything at once. Collapsed state is remembered per browser, so the shape you
+arrange survives a refresh (and the self-heal reload that follows most edits).
+
+While a search filter is active every match is shown regardless of collapsed
+ancestors — otherwise searching for something inside a folded subtree would
+silently return nothing. Clearing the box restores your saved shape.
+
+<a href="images/directory.png" target="_blank"><img src="images/directory.png" alt="Directory & inventory list view" width="80%"></a>
+
+## Slug conventions
+
+Slugs are the stable identifiers automation keys off, so the tooling around the SSO Manager follows a shared convention:
+
+- **Sites**: `site_<name>` — e.g. `site_local`, `site_us-east`
+- **Hosts**: `host_<hostname>` — e.g. `host_pve1`, `host_web01`
+- **Services/apps**: a plain slug or `app_<name>` — e.g. `sso-manager`, `app_emby`
+
+The auto-created LDAP groups derive from the slug (`<slug>_access` / `<slug>_admin`), so keep slugs stable once access groups are in use.
+
+## Automatic registration
+
+You don't have to build the graph by hand — the theta42 tooling registers itself:
+
+### The stack itself (theta-env)
+
+[theta-env](https://github.com/theta42/theta-env)'s `./setup.sh` seeds the directory on every run with the stack it deploys:
+
+- a **site** (name from `CFG_SITE_NAME` in `setup.env`, default `local` → slug `site_local`) marked as the current site
+- the **host** the stack runs on (`host_<hostname>`), with IP, MAC address, OS, and kernel collected from the machine
+- the **hosts** for the proxy and jump host (`host_theta-proxy`, `host_theta-jump`)
+- the **services** it composes — SSO Manager, Proxy (management UI), OpenLDAP Directory (the LDAPS endpoint Linux hosts and LDAP-native apps bind to), OpenResty Edge (the 80/443 data plane), and the SSH Jump Host — each with its address, internal port, and git repo
+- the proxy's auto-registered **OAuth client**, linked under its service
+
+Services are parented to the host that actually runs them: Proxy and OpenResty
+Edge under `host_theta-proxy`, the SSH Jump Host under `host_theta-jump`, and the
+rest under the stack host. Installs seeded before this was fixed had all of them
+under the stack host, leaving the two purpose-made host resources childless; the
+seed re-parents those on its next run, and only when the current parent is the
+one the old code set, so a layout you arranged deliberately is left alone.
+
+The seed is idempotent and non-destructive: a resource whose slug already exists is considered operator-owned — the seed only fills in metadata fields you haven't set, and never overwrites your values.
+
+### Linux hosts (ldap-client)
+
+The `ldap-client` join script enrolls a Debian/Ubuntu machine for LDAP login (SSSD/PAM), LDAP-backed `sudo`, and SSH keys from the directory — and, when given an SSO API token, registers the machine as a `host_<hostname>` resource with its IP, MAC, OS, and kernel, parented to the site named by its configured location.
+
+## Consumers of the directory
+
+The inventory graph isn't just documentation — other components read it to make decisions:
+
+- **[Jump Host](https://theta42.github.io/jump-host/)** — an SSH jump host that resolves which downstream machines a user may reach from their LDAP groups × the directory's `host` resources (`GET /api/discovery/resources?group=<cn>`), then bridges them in. The `host_<hostname>` slugs and `host_<slug>_access` groups this directory creates are exactly what it keys off; a host's `metadata.ip` / `metadata.sshPort` tell it where to connect. So a machine registered here (by theta-env or ldap-client) becomes reachable through the jump host the moment a user is in its access group.
+
+Planned consumers (end-user catalog, firewall/DNS generation) and the model/API gaps they need are tracked in [`directory_spec.md`](https://github.com/theta42/sso-manager-node/blob/master/directory_spec.md) §9.
+
+## Subtype Management & Metrics Drivers Architecture
+
+The Directory includes a **4-tier Driver Resolution Engine** (`services/driver_registry.js`) that binds a resource's `subType` metadata to specific operational protocols for real-time telemetry, log streaming, and remote lifecycle management:
+
+1. **Direct Agent Execution** (`ThetaAgentDriver`): Used when a `theta-agent` daemon is connected to the resource (`systemd`, `docker`, `zfs_pool`, `desktop_linux`, `openrc`, `wireguard`).
+2. **Specialized Subtype Drivers**:
+   - `ProxmoxDriver`: Proxmox VE hypervisors & `lxc` / `kvm` guest controls.
+   - `DockerSocketDriver`: Docker Engine API & `docker_compose` stacks.
+   - `DbDriver`: `postgresql`, `redis`, `openbao_vault`.
+   - `NetworkDriver`: `wireguard`, `unifi_ap`, `unifi_switch`, `pfsense`.
+   - `K8sDriver`: `k8s_pod`, `k8s_deployment`.
+3. **Ancestor / Hypervisor Provider Fallback**: If an LXC/KVM guest lacks a direct agent, the engine automatically queries its parent Proxmox hypervisor node for VMID telemetry and power controls.
+4. **Unmanaged Fallback**: Reports unmanaged status cleanly.
+
+### Subtype Operations API
+- `GET /api/directory-admin/resources/:id/driver-metrics` — Real-time telemetry payload
+- `POST /api/directory-admin/resources/:id/driver-action` — Execute management actions (`{ action, params }`)
+- `GET /api/directory-admin/resources/:id/driver-logs` — Tail operational log output (`?lines=100`)
+
+## Explicit Secret Inheritance Mode
+
+Resource secrets stored in OpenBao (`secret/data/resources/<slug>/conf`) use **Explicit Secret Inheritance Mode** with strict upward ancestor lineage:
+
+- **Strict Ancestor Lineage**: When viewing candidate secrets for inheritance, the dropdown strictly filters to **direct upward ancestors** in the directory hierarchy (Resource $\rightarrow$ Parent Host $\rightarrow$ Cluster $\rightarrow$ Site). Sibling resources across the directory are never exposed.
+- **Explicit Assignment**: Secret pointers (`INHERIT:<parentSlug>:<parentKey>`) are explicitly saved per resource, guaranteeing precise secret scoping across hosts, LXC/KVM containers, and services.
+
+## API
+
+All of the above uses the same admin API the UI does (group `app_sso_directory_admin` or `app_sso_admin`):
+
+- `GET/POST /api/directory-admin/resources`, `PUT/DELETE /api/directory-admin/resources/:id`
+- `GET/POST/DELETE /api/directory-admin/edges` — parent/child links (`hosts`, `oauth` relations)
+- `GET/POST/DELETE /api/directory-admin/groups` — resource ↔ LDAP group links
+- `GET /api/directory-admin/access-summary` — per-resource group + member counts (the Access column)
+- `GET /api/directory-admin/user-access/:uid` — the reverse lookup: every resource a given user can reach, and via which group
+- Read-only graph views (any authenticated user): `GET /api/discovery/resources`, `/api/discovery/resources/:slug`, `/api/discovery/graph`, `/api/discovery/me`
+
+Access requests are open to any authenticated user; deciding is gated per-resource inside the router (resource owner or directory admin):
+
+- `POST /api/access-requests` — `{slug | resourceId, groupCn?, note?}`
+- `GET /api/access-requests/mine` — the caller's own history
+- `GET /api/access-requests` — pending requests the caller may decide
+- `POST /api/access-requests/:id/approve` · `POST /api/access-requests/:id/deny`
+- `DELETE /api/access-requests/:id` — the requester withdraws their own pending request

docs/discovery.md

@@ -0,0 +1,117 @@
+---
+layout: default
+title: Discovery & Inventory
+nav_order: 6
+---
+
+# Discovery & Inventory
+
+[← Back to Home](index.html)
+
+The Directory holds two different kinds of thing, and the distinction matters
+for every consumer of the directory:
+
+- **Catalog resources** — what you have declared. Created by hand, seeded by
+  `setup.sh`, or *promoted* from a discovery result. These get LDAP access
+  groups, appear in the Catalog, and are the only hosts the
+  [jump host](https://github.com/theta42/jump-host) will connect you to.
+- **Discovered resources** — what the network reports. Produced by
+  [discovery plugins](plugins.html) and shown on the **Discovered Inventory**
+  tab. They are a queue of "this exists, do you want to manage it?", not
+  infrastructure you have committed to.
+
+A resource is discovery-only when its `metadata.discovery_sources` is non-empty
+and it has never been promoted. Promoting sets `metadata.managed = true`, at
+which point it becomes catalog content like any other resource.
+
+> Nothing grants access to a discovered resource. It carries no groups until it
+> is promoted, and the jump host applies the same rule — an unpromoted Proxmox
+> guest is not a jump target.
+
+---
+
+## Where discovered data comes from
+
+| Source | What it reports |
+| :--- | :--- |
+| [Proxmox](plugins.html) | The cluster endpoint, its nodes, and every VM/LXC with NICs, `vmid` and node |
+| [UniFi](plugins.html) | Network devices and connected clients, by MAC |
+| [nmap](plugins.html) | Hosts and open ports on a target range |
+| [Docker](plugins.html) | Containers on a local or remote daemon |
+| [theta-agent](agents.html) | The host it runs on — OS, kernel, CPU, RAM, disk, addresses |
+| [ldap-client](directory.html) | A Linux host registering itself when it joins |
+
+An agent is the most authoritative of these: it runs *on* the machine it
+describes. A network scan is the least — it only knows what answered.
+
+---
+
+## How results are matched to existing resources
+
+Every source runs through one reconciler, so two sources seeing the same
+machine converge on one resource instead of creating duplicates. Matching is
+tried in order of precision:
+
+1. **MAC address** — the strongest signal, compared across every interface.
+2. **IP address** — any address on any interface, plus `metadata.address`.
+3. **Slug, name, or base hostname** — last resort.
+
+A candidate must also be **the same kind**. Without that guard a discovered VM
+named `gitea-runner` would match a hand-created *service* of the same name on
+rule 3 and overwrite it. (`template` counts as `host`: converting a VM to a
+template is the same machine.)
+
+When a match is found the metadata is merged, interfaces are unioned by MAC, and
+the source is added to `discovery_sources` — so a resource can legitimately read
+`["unifi", "proxmox"]`, meaning two independent sources agree it exists.
+
+### Naming
+
+Sources disagree about names, so the most human one wins: a **hostname** beats
+an **IP-shaped** name, which beats a **MAC-shaped** name; length is only a
+tie-break within a rank. This is why a device UniFi knows only as
+`ac:16:2d:b3:da:80` is renamed `dl380-0` once Proxmox reports it.
+
+### Relationships
+
+Plugins emit edges as well as resources (a Proxmox node under its cluster
+endpoint, a guest under its node). The reconciler refuses any edge that would
+make a resource its own parent, or that would close a loop — a cycle renders as
+an infinitely nested tree and breaks every ancestor walk in the app.
+
+---
+
+## Promoting a discovered resource
+
+On the **Discovered Inventory** tab, press **Promote**. The resource form opens
+pre-filled with what was discovered — name, kind, address, subtype — so you can
+correct it before committing. Saving marks it managed and provisions its
+[LDAP groups](groups.html).
+
+Each row shows what the directory knows about the device: its source(s), its
+`vmid` where applicable, the identifier it has at that source (`sourceId`, e.g.
+`dl380-0/qemu/234`), and every interface with its MAC and address. If a row
+looks wrong, that detail is where to start.
+
+---
+
+## Stale results
+
+Resources that are *only* auto-discovered are garbage-collected: if a source
+stops reporting one for long enough it is marked
+`lifecycle_state: "archived"` rather than deleted. Anything you created or
+promoted is never touched — `manual` in `discovery_sources` exempts it.
+
+A Proxmox node that is powered off is still reported (with its `status`), so
+downtime does not look like decommissioning.
+
+---
+
+## What the stack discovers about itself
+
+`setup.sh` seeds its own components as catalog resources — the site, the stack
+host, `theta-proxy` and `theta-jump`, and the services under them. The Docker
+discovery plugin then finds the containers backing them. Containers belonging to
+the theta-suite compose project are recognised and attached to the service they
+implement rather than appearing as unmanaged strangers, so a fresh install has an
+empty Discovered Inventory rather than five things demanding attention.

docs/groups.md

@@ -0,0 +1,312 @@
+---
+layout: default
+title: Group & Permission Model
+nav_order: 3
+---
+
+# Theta42 Group & Permission Model
+
+This is the canonical reference for how **groups and permissions work** across the
+theta42 suite (SSO Manager, Proxy, Jump-Host) and how **downstream apps and Linux
+hosts** should read and use them. It is written to be implementable by both humans
+and LLM agents.
+
+Everything below assumes LDAP is the single source of truth for identity and group
+membership. Group membership is managed in the **SSO Manager Directory**, generated
+from adopted resources — there is **no standalone "Groups" page**.
+
+---
+
+## 1. Principles
+
+1. **Groups are a projection of the resource graph.** Every adopted host and app
+   in the Directory gets its own groups, auto-created from its identity. Group
+   membership is managed on the resource's modal.
+2. **Two orthogonal resource namespaces: `host` and `app`.** A host administers
+   hosts; an app administers apps. They do not inherit from each other.
+3. **Three levels per resource: `admin`, `access`, and opaque `capability`.**
+   `admin` implies `access`. Capabilities are explicit and never implied by
+   `admin`.
+4. **Multi-site by prefix.** Each site's groups are fully independent, scoped by
+   the site slug.
+5. **Hosts map, LDAP stays clean.** Directory groups are `groupOfNames` (RBAC)
+   with **no `gidNumber`**. A Linux host uses SSSD to import only the groups it
+   needs and generate their GIDs on the fly (see §8) — no mass import, no GID
+   bloat. Only the meta groups are never imported by hosts.
+6. **The directory is the only place groups are created.** `god_admin` is the sole
+   group that does not belong to a resource or site.
+
+---
+
+## 2. Group schema
+
+`S` = site slug (see §7 for normalization). `<host>`/`<app>` = the resource slug.
+`<capability>` = an opaque, app-defined capability token (see §4).
+
+| Group | Scope | Meaning |
+| :--- | :--- | :--- |
+| `god_admin` | global | **Everything, everywhere** (all sites, hosts, apps, consoles, all capabilities). The only non-site group. |
+| `S_super_admin` | site | Everything on site `S` (all hosts, apps, consoles, all capabilities at `S`). |
+| `S_hosts_admin` | site | Admin on **all hosts** at `S`. |
+| `S_hosts_access` | site | Access to **all hosts** at `S`. |
+| `S_hosts_<capability>` | site | Capability `<capability>` on **all hosts** at `S`. |
+| `S_host_<host>_admin` | host | Admin on host `<host>`. |
+| `S_host_<host>_access` | host | Access to host `<host>`. |
+| `S_host_<host>_<capability>` | host | Capability `<capability>` on host `<host>`. |
+| `S_apps_admin` | site | Admin on **all apps** at `S`. |
+| `S_apps_access` | site | Access to **all apps** at `S`. |
+| `S_apps_<capability>` | site | Capability `<capability>` on **all apps** at `S`. |
+| `S_app_<app>_admin` | app | Admin on app `<app>`. |
+| `S_app_<app>_access` | app | Access to app `<app>`. |
+| `S_app_<app>_<capability>` | app | Capability `<capability>` on app `<app>`. |
+
+### Meta groups (implicit membership — not POSIX, no gidNumber)
+
+| Group | Scope | Meaning |
+| :--- | :--- | :--- |
+| `everyone` | global | **All authenticated users**, any site. |
+| `S_everyone` | site | **All authenticated users** at site `S`. |
+
+These are resolved by the directory (any authenticated user passes), never
+enumerated as LDAP members, and cannot be used as Unix groups.
+
+---
+
+## 3. Naming, normalization & reserved rules
+
+- The **structural delimiter is `_`**. It appears only between the fixed segments
+  of a group name.
+- **Site, host, and app slugs never contain `_`.** Normalize to lowercase;
+  spaces and `_` → `-`; strip other non-`[a-z0-9-]`. A host named `Web 01` and a
+  site `Main Office` produce slugs `web-01` and `main-office`.
+- **Aggregate groups use the plural kind** (`hosts`, `apps`); per-resource groups
+  use the singular (`host`, `app`). This makes `S_hosts_admin` unambiguous even
+  if a host were named `admin` (that host would be `S_host_admin_admin`).
+- **The last segment is the level.** If it is `admin` or `access` it is a known
+  level; any other value is an **opaque capability** owned by a downstream app.
+- **Total length budget:** keep a group cn under ~120 chars; reject group
+  creation that would exceed it.
+- Groups are **`groupOfNames`** (RFC 2307bis) with **no `gidNumber`**. GIDs are
+  generated on the host by SSSD for only the groups that host imports (see §8).
+
+---
+
+## 4. Levels and opaque capabilities
+
+- **`admin`** — manage (create/update/delete/config) the resource.
+- **`access`** — use/read the resource.
+- **`<capability>`** — an arbitrary token the SSO does **not** interpret. The SSO
+  manages membership and exposes the group to the app; **the downstream app
+  defines and enforces what the capability means** (e.g. `emby_admin`,
+  `gitea_maintain`, `reboot`, `backup`).
+
+The directory recognizes `admin`, `access`, `super_admin`, and the meta groups.
+Everything else on a resource group is treated as an opaque capability group and
+passed through to consumers.
+
+---
+
+## 5. Permission resolution (inheritance)
+
+Define a user's **effective permission** on a resource by checking, from most
+specific to most general, whether they are a member of any applicable group. The
+rule: a higher group implies everything below it.
+
+### On host `H` at site `S`
+
+| Wanted | Granted if the user is a member of **any** of |
+| :--- | :--- |
+| **admin** on `H` | `god_admin` · `S_super_admin` · `S_hosts_admin` · `S_host_H_admin` |
+| **access** on `H` | (any admin rule above) · `S_hosts_access` · `S_host_H_access` |
+| **capability `C`** on `H` | `god_admin` · `S_super_admin` · `S_hosts_C` · `S_host_H_C` |
+
+### On app `A` at site `S`
+
+Identical, with `app`/`apps` substituted for `host`/`hosts`.
+
+### Management console (SSO / Proxy / Jump-Host)
+
+Each console is registered as an **app** on its site, so console admin is:
+
+`god_admin` · `S_super_admin` · `S_app_<console>_admin`
+
+### Pseudocode
+
+```
+def effective(resource, level_or_cap, site):
+    if user in "god_admin":                  return True
+    if user in f"{site}_super_admin":        return True
+    if level_or_cap in ("admin","access"):
+        agg = f"{site}_{resource.kind}s_{level_or_cap}"
+        if user in agg:                      return True
+    specific = f"{site}_{resource.kind}_{resource.slug}_{level_or_cap}"
+    if user in specific:                     return True
+    if level_or_cap == "access":             return effective(resource, "admin", site)
+    if level_or_cap == "admin":              return False          # access does not imply admin
+    return False
+```
+
+`everyone` / `S_everyone` are a special grantee: if a resource grants a group to
+`everyone` (or `S_everyone`), any authenticated user (at that site) passes.
+
+---
+
+## 6. Where groups live — the Directory, generated from adopted resources
+
+- There is **no standalone Groups page.** Group creation/management happens on an
+  **adopted resource** in the Directory.
+- When a host or app is **adopted** (promoted from Discovered Inventory to
+  managed), the directory auto-creates its `_admin` and `_access` groups (and
+  site aggregates if configured). Capability groups are created on demand.
+- Membership (add/remove users) and capability grants are managed on that
+  resource's modal.
+- Deleting a resource removes its per-resource groups.
+- The `S_super_admin`, `S_hosts_*`, `S_apps_*`, `S_everyone` site groups and the
+  global `god_admin`/`everyone` are managed at the site level (not on a single
+  host/app resource).
+
+---
+
+## 7. Multi-site isolation
+
+One LDAP tree can serve many sites ("Main Office", "Branch Office", "co-lo",
+"Mikes Homelab", …). Each site `S` has its own fully independent set of `S_*`
+groups behind its prefix. A `main-office_super_admin` or `main-office_hosts_admin`
+touches nothing in `branch-office_*` or `steves-homelab_*`. Only `god_admin` and
+`everyone` cross site boundaries.
+
+---
+
+## 8. Unix/POSIX groups — mapped on the host, not in LDAP
+
+Directory groups are **`groupOfNames`** (RFC 2307bis) and carry **no `gidNumber`**.
+There are hundreds of them and only a handful matter on any given host, so we do
+**not** bloat LDAP with GIDs. Instead, each Linux host uses SSSD to import only the
+groups it cares about and map them to GIDs **on the fly** (algorithmic ID mapping).
+This keeps the directory clean and the per-host surface tiny.
+
+### SSSD — generate GIDs on the fly, import only what you need
+
+```ini
+[domain/example]
+id_provider = ldap
+auth_provider = ldap
+ldap_uri = ldaps://ldap.example
+ldap_search_base = dc=example,dc=com
+
+# groupOfNames (RFC 2307bis) schema
+ldap_schema = rfc2307bis
+ldap_group_object_class = groupOfNames
+ldap_group_member = member
+
+# Map GIDs mathematically from the LDAP UUID — no gidNumber in LDAP
+ldap_id_mapping = true
+ldap_group_uuid = entryUUID
+
+# Import ONLY the groups this host needs (e.g. a naming convention or an OU)
+ldap_group_search_filter = (&(objectClass=groupOfNames)(cn=linux-*))
+```
+
+Key ideas:
+- `ldap_id_mapping = true` + `ldap_group_uuid = entryUUID` make SSSD derive a
+  stable GID for any group it imports, so **no `gidNumber` attribute is required**
+  in LDAP.
+- `ldap_group_search_filter` is the gatekeeper: SSSD imports only groups that
+  match, discarding the other hundreds. After changing the filter, clear the
+  cache (`sss_cache -E`; `rm -f /var/lib/sss/db/*`; restart sssd) and verify with
+  `getent group <cn>`.
+
+### What filter to use — the naming convention is the answer
+
+A host should import its **own** resource groups (plus any explicitly granted
+ones). Because the schema is predictable, `ldap-client` can generate the per-host
+`ldap_group_search_filter` from the enrolled host's identity, e.g. a host `web01`
+at site `main-office` imports:
+
+```
+(&(objectClass=groupOfNames)(|(cn=main-office_host_web01_access)
+                               (cn=main-office_host_web01_admin)
+                               (cn=main-office_host_web01_sudo)))
+```
+
+So the operator (or ldap-client) selects a small allowlist of the host's `_access`
+/ `_admin` / capability groups to feed sudoers, SSH `AllowGroups`, and filesystem
+ACLs. **Only those groups are imported** — no GID bloat, no mass import.
+
+### Aliasing an LDAP group into a local group (e.g. `input`)
+
+SSSD cannot merge an LDAP group into a local group whose GID varies per host.
+Two host-side mechanisms cover it:
+
+- **pam_exec** — a script in the login stack adds the user to the local group for
+  the session:
+  ```sh
+  #!/bin/bash
+  if id -Gn "$PAM_USER" | grep -q "host_input"; then usermod -a -G input "$PAM_USER"; fi
+  ```
+  `session optional pam_exec.so /usr/local/bin/add_to_input.sh` in
+  `/etc/pam.d/common-session`.
+
+- **nss-groupmerge** — merge an LDAP group into a local group at NSS time
+  (`/etc/groupmerge.conf`: `input: host_input`, then `group: files sssd groupmerge`
+  in `/etc/nsswitch.conf`), so any service querying `input` sees the LDAP group's
+  members regardless of the local GID.
+
+### Meta groups
+
+`god_admin`, `everyone`, and `S_everyone` are NOT imported by hosts — they have
+implicit membership and are resolved by the directory only.
+
+---
+
+## 9. Downstream-app consumption guide
+
+A downstream app (Emby, Gitea, a custom service, a shell script) reads group
+membership from LDAP and interprets it as follows:
+
+1. **Discover the user's groups** — bind with the user's credentials (or use a
+   service account + `memberOf`). Groups are `groupOfNames` (member DN), so query
+   by the user's DN, e.g. `(&(objectClass=groupOfNames)(member=<user_dn>))`, or use
+   the `memberOf` reverse attribute on the user's entry.
+2. **Match each group to a scope:**
+   - `god_admin` → the user is a global administrator.
+   - `{site}_super_admin` → site administrator for that site.
+   - `{site}_hosts_*` / `{site}_app_*` (aggregate) → applies to all hosts/apps at the site.
+   - `{site}_host_<host>_*` / `{site}_app_<app>_*` → applies to that one resource.
+   - `everyone` / `{site}_everyone` → the user is implicitly a member.
+3. **Interpret the last segment:**
+   - `admin` → full control of that resource.
+   - `access` → read/use.
+   - anything else → a capability **you** define; act on it or ignore it.
+4. A user with `{site}_host_web01_access` can reach `web01`; a user with
+   `{site}_host_web01_reboot` (if you define `reboot`) may reboot it; a user with
+   `{site}_app_emby_emby_admin` administers Emby.
+
+The app must **never** treat an unknown last segment as `admin` or `access`.
+
+---
+
+## 10. Migration from the legacy `app_*` groups
+
+The current global groups (`app_sso_admin`, `app_super_admin`,
+`app_sso_directory_admin`, `app_jump_admin`) are replaced by the new model:
+
+| Legacy | New |
+| :--- | :--- |
+| `app_super_admin` | `god_admin` |
+| `app_sso_admin` | `S_app_sso_admin` (+ `S_super_admin` for site admins) |
+| `app_sso_directory_admin` | `S_app_sso_admin` |
+| `app_jump_admin` | `S_app_jump_admin` |
+
+During the transition the legacy groups may be kept as short-lived aliases that
+resolve to the same effective permission; once everything is moved, remove them.
+
+---
+
+## 11. The management consoles are apps
+
+The SSO, Proxy, and Jump-Host each register themselves as an app on their site and
+receive their auto-generated groups (`S_app_sso_admin`, `S_app_proxy_admin`,
+`S_app_jump_admin`, plus `_access`). Their admin UIs gate on
+`god_admin` · `S_super_admin` · `S_app_<console>_admin`. This keeps everything
+self-consistent: the SSO is "just another app."

docs/index.md

@@ -1,6 +1,7 @@
 ---
 layout: default
 title: Home
+description: A self-hosted OpenID Connect provider with a bundled OpenLDAP directory and a web management UI. One login for your modern apps, one LDAP directory for the rest, no phone-home.
 ---
 
 # SSO Manager
@@ -21,10 +22,11 @@ one command).
 
 ## Screenshots
 
-<a href="images/dashboard.png" target="_blank"><img src="images/dashboard.png" alt="Dashboard" width="49%"></a>
+<a href="images/dashboard.png" target="_blank"><img src="images/dashboard.png" alt="Overview dashboard" width="49%"></a>
 <a href="images/users.png" target="_blank"><img src="images/users.png" alt="User list" width="49%"></a>
 <a href="images/groups.png" target="_blank"><img src="images/groups.png" alt="Groups" width="49%"></a>
-<a href="images/oauth-clients.png" target="_blank"><img src="images/oauth-clients.png" alt="OAuth clients" width="49%"></a>
+<a href="images/directory.png" target="_blank"><img src="images/directory.png" alt="Directory & inventory" width="49%"></a>
+<a href="images/oauth-clients.png" target="_blank"><img src="images/oauth-clients.png" alt="OAuth client (edit view)" width="49%"></a>
 
 *(click any screenshot to view full size)*
 
@@ -51,10 +53,17 @@ backend, that's the niche.
 - **Web management UI** — users, groups, and OAuth clients from a browser;
   invite and password-reset flows over email; self-service profile + API
   tokens.
-- **LDAPS for legacy apps** — anything that binds LDAP directly (Gitea,
-  Emby, …) uses LDAPS/StartTLS against the same directory.
+- **Direct LDAP binds** — anything that binds LDAP directly (Linux hosts
+  via PAM/SSSD, Gitea, Emby, …) uses LDAPS/StartTLS against the same
+  directory.
 - **All-in-one Docker image** — app + OpenLDAP + Redis in one container, or
   run the pieces separately via `app_*` env config.
+- **Geo-Location Scaling** — built-in support for N-Way Multi-Master OpenLDAP [replication](replication.html) across physical sites.
+- **[Directory & Inventory](directory.html)** — map sites, hosts, and services as a graph with rich metadata (IP/MAC, OS/kernel, ports, git repos), auto-provisioned access groups, and automatic registration from theta-env and ldap-client. Drives directory-aware tools like the [SSH jump host](https://theta42.github.io/jump-host/).
+- **[Discovery](discovery.html)** — the catalog-vs-discovered distinction, how scanned assets are matched/merged into existing resources, and how a discovery gets promoted into the catalog (and becomes reachable through the jump host).
+- **[Theta Agent & Endpoint C2](agents.html)** — 2-way Go daemon (`theta-agent`) for real-time telemetry (CPU, RAM, Disk, ZFS, GPU), automated host discovery, SSSD/LDAP configuration, and local capability-controlled management operations.
+- **[Vault secrets](vault.html)** — an OpenBao-backed key-value store built into the UI, for stashing passwords/API keys/credentials with encryption and access control.
+- **[API tokens](concepts-api-tokens.html)** — self-service personal access tokens for calling the management API from scripts/CI without a browser session.
 
 ## Get it
 
@@ -74,5 +83,7 @@ That's the standalone quick start. For the full set of install options
 
 - **[Proxy](https://theta42.github.io/proxy/)** — an OIDC + LDAP-aware
   reverse proxy, designed to sit in front of this SSO.
+- **[Jump Host](https://theta42.github.io/jump-host/)** — an SSH jump host that
+  uses this SSO's directory to decide who may reach which machine.
 - **[theta-env](https://theta42.github.io/theta-env/)** — runs this SSO
   Manager and the proxy together with one command.

docs/ldap.md

@@ -1,16 +1,22 @@
 ---
 layout: default
 title: LDAP
+description: SSO Manager's bundled OpenLDAP directory — schema, service accounts, TLS, and connecting third-party apps directly.
 ---
 
 # LDAP Directory
 
 [← Back to Home](index.html)
 
+> Looking for a plainer explanation of accounts, groups, and managers
+> instead of schema/attribute detail? See
+> [Accounts, Groups & Managers](concepts-accounts.html).
+
 SSO Manager runs an OpenLDAP directory holding your users and groups. The app
 authenticates against it over `localhost:389` (inside the all-in-one container)
-and exposes **LDAPS** (`ldaps://…:636`, TLS) for legacy apps that bind LDAP
-directly — Gitea, Emby, the theta42/proxy, etc.
+and exposes **LDAPS** (`ldaps://…:636`, TLS) for anything that binds LDAP
+directly — Linux hosts (PAM/SSSD, sudo rules, SSH keys), Gitea, Emby, the
+theta42/proxy, etc.
 
 ## Directory layout
 
@@ -34,6 +40,15 @@ User entries are `cn=<uid>,ou=people,<base>` and carry the objectClasses:
 - `sudoRole` — per-user sudo rules (`sudoCommand`, `sudoHost`, `sudoUser`).
 - `theta42Person` (custom auxiliary; `dateOfBirth`).
 
+Every user (person or service account) also carries a `manager` attribute
+(the standard COSINE `manager`, `SUP distinguishedName`) — one or more DNs of
+the people who created/administer that account. Set automatically to the
+creator's DN on signup (whoever an admin was logged in as, or whoever sent
+the invite), and reassignable later from the account's Edit form. Anyone
+listed as a `manager` can edit that account (same fields an admin can:
+mobile, description, SSH key, date of birth, home directory, login shell,
+and the manager list itself) without needing `app_sso_admin`.
+
 Passwords are stored as `{SSHA512}` (8-byte salt, sha512(pass+salt), base64),
 verified by the `pw-sha2` module. The app's `hashPasswordSSHA512` is the
 canonical hasher; if you provision users out-of-band, hash passwords the same
@@ -44,18 +59,64 @@ way or use `slappasswd -h '{SSHA512}'`.
 Groups are `cn=<name>,ou=groups,<base>` (`groupOfNames`) with a `member`
 attribute listing member DNs. The `memberOf` overlay populates reverse
 membership (`memberOf` on the user); `refint` keeps it consistent on
-add/remove. **Admin permission checks read the group's `member` list**, not
-`memberOf` on the user.
+add/remove.
 
-The SSO requires three groups (seeded automatically by the entrypoint /
-`install.sh`):
+Note that `groupOfNames` requires **at least one member**, which has two
+consequences worth knowing: whoever creates a group is automatically seeded
+into it, and removing the last member (user *or* nested group) is refused with
+a 409 rather than leaving an invalid entry behind.
+
+### Nested groups
+
+A `member` DN may be another group's, not just a user's — that is how nesting
+is stored, with no extra schema. Everyone in the nested group is a member of
+the outer one, at any depth. Manage it on the **Groups** page under each
+group's *Nested* tab, or via the API:
+
+```
+PUT    /api/group/:group/nested/:child    nest :child inside :group
+DELETE /api/group/:group/nested/:child    un-nest
+GET    /api/group/:group/effective        direct users, nested groups, and the
+                                          full transitive set of users
+```
+
+Cycles are refused (409) rather than truncated — a loop makes "who is in this
+group" unanswerable. Two standing relationships are wired automatically: the
+cross-app `app_super_admin` is nested into every resource's `<slug>_admin`
+group, and each `<slug>_admin` into its `<slug>_access` group, so administering
+something implies being able to use it.
+
+**Resolving nesting is a client-side job on stock OpenLDAP.** No 2.6.x release
+can evaluate nested groups; `memberOf` and a `(member=X)` filter both return
+direct membership only. The bundled slapd is therefore built from source with
+the `nestgroup` overlay (see *Modules + overlays* below), and the app is told so
+via `ldap.nestedGroupsServerSide`. Against any other server the app computes the
+closure itself — same answers, more queries. Either way, **never read `memberOf`
+directly to make an access decision**; use `utils/user_groups.js`'s `groupCns()`,
+which is correct in both modes.
+
+### Personal groups
+
+Every user (person or service account) also gets a **personal Unix group**
+at creation — `cn=<uid>,ou=groups,<base>`, `objectClass: posixGroup` (RFC
+2307), holding just `cn` and `gidNumber` (the user's primary GID). This is a
+different schema than the `groupOfNames` groups above — its membership
+attribute is `memberUid` (a bare username, not a DN), and unlike
+`groupOfNames` it's valid with zero members. It's excluded from the
+`/groups` page (which filters on `objectClass=groupOfNames`) and managed
+instead from the owning user's own profile page ("Members of `<uid>`'s
+group", admin-only) — add other accounts as supplementary members, e.g. to
+share write access to files owned by this group.
+
+The SSO seeds these groups automatically (entrypoint / `install.sh`):
 
 | Group | Grants |
 |-------|--------|
+| `app_super_admin` | cross-app super admin. Nested into the three below, so its members hold those rights transitively rather than by a special case in app code — and the privilege is visible to LDAP-native consumers (SSSD, sudo) too. |
 | `app_sso_admin` | full admin (users, groups, settings) |
 | `app_sso_oauth_admin` | OAuth client management |
 | `app_sso_invite` | invitation management |
-| `app_sso_service_account` | not a permission — marks a `posixAccount` as a non-person service account (see *Service accounts* below) |
+| `app_sso_service_account` | not a permission — marks a `posixAccount` as a non-person service account (see *Service accounts* below). Deliberately **not** nested into, since it changes how an account is displayed rather than what it may do. |
 
 ## TLS (LDAPS / StartTLS)
 
@@ -91,35 +152,116 @@ volumes:
 
 The entrypoint leaves existing certs untouched (idempotent).
 
+## Choosing the LDAPS hostname
+
+The `/integrations` page advertises an **LDAPS URL** for direct LDAP binds. By
+default it derives that URL from the public OAuth issuer (e.g.
+`https://sso.example.com` → `ldaps://sso.example.com:636`). That is convenient,
+but it implies LDAP clients reach your directory through the same public
+hostname — which usually means port-forwarding 636 through your router.
+
+**Do not port-forward LDAPS (636) to the public internet.** LDAP simple binds
+have no rate limiting and are a brute-force target. Instead, use one of these
+internal-only patterns and set `conf.ldap.ldapsHost` (or
+`app_ldap__ldapsHost`) so the `/integrations` page shows the right URL.
+
+### 1. Same Docker / local network host (best for apps on this machine)
+
+If the LDAP client runs on the same Docker network as the SSO Manager (for
+example, the bundled `theta-env` stack), use the internal service name:
+
+```
+ldaps://sso-manager:636
+```
+
+In `conf/secrets.js`:
+
+```javascript
+ldap: {
+  ldapsHost: 'sso-manager',
+  ldapsPort: 636,
+}
+```
+
+The proxy in theta-env already uses this internally. The bundled slapd cert
+includes `sso-manager` in its SAN when `LDAP_CERT_CN` is left at its default,
+so hostname verification works without extra setup.
+
+### 2. LAN host behind your router (best for separate home-lan machines)
+
+Create an internal-only DNS record — e.g. `ldap.internal.example.com` →
+`192.168.1.10` — using your router, Pi-hole, or a local `hosts` file. Then get
+or generate a cert whose SAN/CN matches that internal name:
+
+- **Let's Encrypt wildcard** (`*.internal.example.com`) works if you own the
+  public domain and can complete DNS-01 challenge; the record itself can stay
+  private/routable only inside your LAN.
+- **Internal CA** is fine for a pure LAN: run a small CA, issue a cert for
+  `ldap.internal.example.com`, and distribute the CA cert to clients.
+- **Self-signed** with `LDAP_CERT_CN=ldap.internal.example.com` also works; copy
+  the generated `ldap.crt` to each client and trust it.
+
+In `conf/secrets.js`:
+
+```javascript
+ldap: {
+  ldapsHost: 'ldap.internal.example.com',
+  ldapsPort: 636,
+}
+```
+
+The URL on `/integrations` becomes `ldaps://ldap.internal.example.com:636`.
+
+### 3. Public hostname (acceptable only behind a VPN/firewall)
+
+If a remote host must bind LDAP, put it behind a VPN (Tailscale, WireGuard,
+etc.) or a tightly locked-down firewall rule. In that case the public hostname
+may be appropriate, but the LDAPS port should still not be reachable from the
+open internet.
+
+### Why not just use the LDAP server's IP address?
+
+TLS clients verify the server name against the certificate. Connecting to
+`ldaps://192.168.1.10:636` with a cert issued for `*.internal.example.com`
+will fail hostname verification unless you disable cert checks — which removes
+most of the security benefit of LDAPS. Always use a hostname that matches the
+cert.
+
 ## Service accounts
 
-There are two different kinds of "not a real person" account, and which one
-you want depends on what's consuming it:
+A service account is a normal `posixAccount` for something that isn't a
+person: a media manager, a torrent client, a service like Emby, or a
+read-only bind account an app uses to look users up — anything that needs a
+real `uidNumber`/`gidNumber` to own files, or that other accounts join via a
+group for write access (e.g. a `stuff_manager` group granting write rights
+to a media library). There's only one kind — every account, person or
+service, is a real `posixAccount` with a UID.
 
-**LDAP bind-only** — for an app that just needs to bind LDAP to look users up
-(its own "LDAP authentication" settings page, or the read-only account
-`theta42/ldap-client` binds as). Not a `posixAccount` — no `uidNumber`, no
-home directory, can't log into this UI. Create one from the
-**Integrations → LDAP** tab's *Service Accounts* section (create, rotate
-password, delete). theta-env's bootstrap creates `cn=ldapclient` this same
-way automatically, and the proxy binds as it — don't reuse the admin DN for
-this.
+Create one from the **Users → Service Accounts** tab's "Add new user" form
+with **This is a service account** checked — it skips the birthday/
+Terms-of-Service fields a real person's account needs and asks for just an
+account name. It's flagged (via membership in the `app_sso_service_account`
+group) so it's listed separately from real people and excluded from "all
+users" notification broadcasts.
 
-**Unix/POSIX** — for an account something actually *runs as* on a Linux
-host: a media manager, a torrent client, a service like Emby — anything that
-needs a real `uidNumber`/`gidNumber` to own files or that other accounts join
-via a group for write access (e.g. a `stuff_manager` group granting write
-rights to a media library). Create one from the **Users** page's "Add new
-user" form with **This is a service account** checked — it skips the
-birthday/Terms-of-Service fields a real person's account needs and asks for
-just an account name. It's a normal `posixAccount`, just flagged (via
-membership in the `app_sso_service_account` group) so it's visibly marked in
-the Users list and excluded from "all users" notification broadcasts.
+Email and password are both optional for a service account:
 
-Either way: don't reuse the admin DN, and give it only the group memberships
-it actually needs.
+- No `mail` is set unless you give it one (it never needs a mailbox).
+- Leaving the password blank is fine — no `userPassword` attribute is set at
+  all, and an entry with no `userPassword` simply can't bind with any
+  password (standard LDAP simple-bind behavior). Only set a password if the
+  account actually needs to authenticate as itself (e.g. a bind-only account
+  an app uses to look users up).
 
-Example bind test (LDAP bind-only account):
+theta-env's bootstrap creates its own `cn=ldapclient` bind account directly
+against LDAP (independent of this app), and the proxy binds as it — that
+account won't show up in the Service Accounts tab since it isn't managed
+through this app, but it keeps working unchanged.
+
+Either way: don't reuse the admin DN, and give a service account only the
+group memberships and `manager`s it actually needs.
+
+Example bind test (a service account with a password set):
 
 ```bash
 ldapsearch -x -H ldaps://sso.example.com:636 \
@@ -199,11 +341,37 @@ needs:
 
 - **Modules:** `pw-sha2` (the app stores user passwords as `{SSHA512}`),
   `ppolicy`, `memberof`, `refint`.
+- **Optional — `nestgroup`:** server-side nested-group evaluation. Not in any
+  released OpenLDAP (added to master as ITS#10161 in March 2024; 2.7 is still
+  unreleased), so the bundled image builds slapd from a pinned upstream commit.
+  Without it the app resolves nesting itself and everything still works — leave
+  `ldap.nestedGroupsServerSide` at `false`. With it, set that to `true` and
+  configure:
+
+  ```
+  overlay         nestgroup
+  nestgroup-base  ou=groups,<base>
+  nestgroup-flags member-filter memberof-filter memberof-values
+  ```
+
+  Flags are **space-separated**; the comma form the man page's `{a, b, c}`
+  notation suggests is rejected. `member-values` is deliberately omitted — it
+  expands the `member` attribute when reading a group, which destroys the
+  distinction between "listed here" and "reachable through a nested group", and
+  the raw values are then unrecoverable. Transitive answers come from the filter
+  flags and from `GET /api/group/:group/effective`.
+
+  One more consequence of building from master: it ships **LMDB 1.0.0**, whose
+  on-disk format is mutually unreadable with the 0.9.x in 2.6.x
+  (`MDB_INVALID: File is not an LMDB file`). Moving a directory between the two
+  is a `slapcat` → `slapadd` reload, not a restart.
 - **Custom schema:** the `theta42Person` auxiliary objectClass with
   `dateOfBirth` — see `ops/ldap-setup.sh` for the LDIF.
 - **Directory tree:** `ou=people`, `ou=groups`, `ou=policies` under the base DN,
   a default `pwdPolicy` at `cn=ppolicy,ou=policies,<base>`.
-- **Required groups:** `app_sso_admin`, `app_sso_invite`, `app_sso_oauth_admin`.
+- **Required groups:** `app_sso_admin`, `app_sso_invite`, `app_sso_oauth_admin`,
+  and `app_super_admin` (the cross-app super-admin group; the bundled entrypoint
+  also nests it into the first three).
 
 `ops/ldap-setup.sh -p <admin-password>` configures all of the above
 idempotently against a running slapd (auto-detects the database holding your

docs/oauth.md

@@ -1,12 +1,17 @@
 ---
 layout: default
 title: OAuth / OIDC
+description: SSO Manager's OpenID Connect / OAuth 2.0 provider — discovery document, client registration, and token endpoints.
 ---
 
 # OAuth 2.0 / OpenID Connect
 
 [← Back to Home](index.html)
 
+> Looking for a plainer explanation of clients/scopes/redirect URIs instead
+> of endpoint-level detail? See
+> [Connecting Apps (Single Sign-On)](concepts-oauth-apps.html).
+
 SSO Manager is an **OpenID Connect / OAuth 2.0 provider**: it issues its own
 access, refresh, and ID tokens that your apps can consume to authenticate
 users and authorize API calls. It also runs a full OpenLDAP directory, so it
@@ -48,20 +53,18 @@ An OAuth client represents an app that authenticates against the SSO. Each has:
 
 ### Managing clients
 
-Clients are managed from the web UI (as a member of the `app_sso_oauth_admin`
-group) or the HTTP API at `/api/oauth/client` (auth via the `auth-token` header
-from a login):
+Clients are managed directly from the **Directory** tab in the web UI. They are modeled as resources of `kind: oauth` and must belong to a parent Service.
 
-| Method | Path | Action |
-|--------|------|--------|
-| `GET` | `/api/oauth/client` | list clients |
-| `POST` | `/api/oauth/client` | create a client (returns the raw `client_secret` once) |
-| `GET` | `/api/oauth/client/:id` | get one |
-| `PUT` | `/api/oauth/client/:id` | update redirect URIs / scopes / groups |
-| `DELETE` | `/api/oauth/client/:id` | delete |
-| `POST` | `/api/oauth/client/:id/rotate` | rotate the secret (returns the new raw secret once) |
+| Action | How to do it |
+|--------|--------------|
+| **Create** | Click the green **+** on a parent Service to add a child resource. Choose **OAuth Integration**. The raw `client_secret` is shown once upon creation. |
+| **Edit** | Click the edit pencil on the OAuth resource in the Directory list or tree. You can update redirect URIs, scopes, allowed groups, and token TTLs. |
+| **Delete** | Click the trash can on the OAuth resource in the Directory list. |
+| **Rotate Secret** | Open the edit modal for the OAuth resource and click **Rotate Client Secret**. The new raw secret is shown once. |
 
-> All client-management endpoints are gated by the `app_sso_oauth_admin` group.
+> All client-management actions use the standard Directory API (`/api/directory-admin/resources`) and are gated by the `app_sso_directory_admin` group.
+
+<a href="images/oauth-clients.png" target="_blank"><img src="images/oauth-clients.png" alt="Editing an OAuth client resource" width="80%"></a>
 
 ## Scopes
 

docs/plugins.md

@@ -0,0 +1,184 @@
+# Plugins
+
+The SSO Manager runs **plugins** as scheduled background tasks. A plugin
+**type** is an installed module; a plugin **instance** is a configured, loadable
+copy of a type. You can create, edit, load/unload, run, and delete instances
+from the **Plugins** page (or the `/api/plugins` API), and you can run several
+instances of the same type — e.g. two Proxmox endpoints, each with its own URL
+and token on its own schedule.
+
+Per-instance **secrets** are stored in [OpenBao](https://openbao.org/) at
+`secret/plugins/<instance-id>/conf`, not in `sso-secrets.js`. The admin UI only
+ever shows them masked (`********`); the plugin reads them at run time. This
+needs theta-suite ≥ v1.30.1 (which grants the `sso-broker` OpenBao policy
+`secret/plugins/*`); re-run `./setup.sh` after upgrading.
+
+## Plugin types
+
+A plugin type is a module under `nodejs/plugins/<category>/<type>.js`. The
+filename basename (without `.js`) is the `type`; the parent directory is the
+`category`. Two built-in categories ship today:
+
+**`discovery`** — scheduled scans that sync external assets into the
+directory catalog:
+
+- `proxmox` — Proxmox VE (URL + API token)
+- `unifi` — UniFi Network controller (URL + username/password)
+- `nmap` — nmap OS + port scan (a target range; no credentials)
+- `docker` — Docker daemon discovery (containers as directory resources)
+
+**`messaging`** — on-demand delivery for alerts, 2FA codes, and
+notifications:
+
+- `twilio` — Twilio SMS
+- `webhook` — universal REST webhook (custom JSON payload to Slack, Teams,
+  Discord, or any HTTP endpoint)
+
+If no messaging plugin instance is enabled, the system falls back to the
+legacy `voipms` integration configured directly in the SSO secrets.
+
+### What the Proxmox plugin produces
+
+One endpoint becomes one subtree:
+
+```
+Proxmox endpoint (cluster name, or the endpoint hostname)
+└── node (hypervisor)
+    ├── VM / template
+    └── LXC / template
+```
+
+The endpoint resource stands for the cluster, not a machine, so it carries the
+API URL and a `sourceId` but deliberately no IP — giving it the address it is
+reached at made the reconciler merge it with the node answering on that address,
+which produced a resource that was its own parent.
+
+Every guest carries:
+
+- `interfaces[]` — one entry per NIC with its own `mac`, `ip`/`ips` and `name`.
+  The MAC and the address on it are read from the same source, so they cannot be
+  mismatched (an earlier version collected MACs and IPs into two flat lists and
+  zipped them by index, which attributed addresses to the wrong NIC on any
+  multi-NIC guest).
+- `macAddress` / `ip` — the primary NIC's values, preferring one that actually
+  has an address.
+- `vmid`, `node` and `sourceId` (`<node>/qemu/<vmid>` or `<node>/lxc/<vmid>`), so
+  a directory row traces back to the exact guest on the exact node.
+
+Interfaces belonging to something running *inside* a guest — `docker0`, `veth*`,
+`br-*`, VPN tunnels — are filtered out. They are not NICs of the host, and their
+172.x addresses would otherwise give the reconciler spurious matches.
+
+A stopped VM still reports its MAC (read from the VM config rather than the
+guest agent), and a DHCP-configured LXC gets its address from the running
+container's interface list. Offline nodes are recorded with `status` rather than
+skipped, so a hypervisor that is down does not look decommissioned and get
+garbage-collected after a week.
+
+A module exports a **manifest**:
+
+```javascript
+module.exports = {
+  // Identity — `type`/`category` default to the file/dir name but can be set
+  // explicitly. `name`/`description` show up in the UI.
+  type: 'proxmox',
+  category: 'discovery',
+  name: 'Proxmox VE',
+  description: 'Discover VMs, containers, and nodes from a PVE endpoint.',
+
+  // Drives the admin UI form, API validation, and secret masking. Fields with
+  // `secret: true` are stored in OpenBao; the rest live in the DB row.
+  configSchema: [
+    { key: 'url',         label: 'API URL',       type: 'url',      required: true },
+    { key: 'tokenId',     label: 'Token ID',      type: 'text',     required: true },
+    { key: 'tokenSecret', label: 'Token Secret',  type: 'password', required: true, secret: true }
+  ],
+
+  // "Test" button: validate the config (don't do the work). Return
+  // { ok: true } or { ok: false, error: '...' }. Optional.
+  validate: async (config) => { … },
+
+  // The work. `run` is the generalized contract name; the discovery plugins
+  // also keep `discover` as an alias for back-compat. For `category:
+  // 'discovery'`, the scheduler passes the result to the discovery reconciler.
+  run: async (config) => { return { resources, edges }; },
+  discover: async (config) => { return { resources, edges }; }
+};
+```
+
+`run(config)` receives the merged non-secret config + secret values as one flat
+object (e.g. `{ url, tokenId, tokenSecret }`). For a discovery plugin it
+returns `{ resources, edges }`; the reconciler upserts them into the resource
+graph attributed to the instance's **slug** (the `discovery_sources` name).
+
+### Writing a custom plugin type
+
+Drop a `.js` file under `nodejs/plugins/discovery/` (or a new category directory)
+following the manifest above. New types are picked up at boot, so restart the
+SSO Manager after adding one. Runtime load/unload is per-**instance** only —
+adding a new type still needs a restart.
+
+## The Plugins page
+
+Under **Plugins** (nav, admin-only — `app_sso_admin` / `app_sso_directory_admin`
+/ `app_super_admin`):
+
+- **New Plugin** — pick a type, name it, choose a unique slug (the discovery
+  source name + the URL the resource graph attributes results to), set a cron
+  schedule, and fill in the config form (secret fields are password inputs).
+  Creating it schedules it and kicks one immediate run.
+- **Edit** — name, cron, and non-secret config.
+- **Edit Secrets** (key icon) — password fields, prefilled masked. Leave a
+  field blank to keep its current value.
+- **Test** (vial icon) — runs the plugin's `validate`.
+- **Run now** (play icon) — enqueues one immediate run regardless of state.
+- **Load / Unload** — enable/disable the schedule without deleting the instance.
+- **Delete** — removes the schedule, the OpenBao secret namespace, and the row.
+
+## API
+
+All endpoints are mounted at `/api/plugins`, require an authenticated admin
+(`app_sso_admin` / `app_sso_directory_admin` / `app_super_admin`), and return
+secret values masked.
+
+| Method + path | Purpose |
+|---|---|
+| `GET /api/plugins/types` | list installed plugin types + their `configSchema` |
+| `GET /api/plugins` | list instances (with masked secrets + last-run state) |
+| `GET /api/plugins/:id` | one instance |
+| `POST /api/plugins` | create — body `{ pluginType, name, slug, cron, config }` where `config` is a flat object of all field values; secret fields are split into OpenBao |
+| `PUT /api/plugins/:id` | update name/cron/enabled + non-secret config |
+| `PUT /api/plugins/:id/secrets` | update secret fields (blank = keep) |
+| `POST /api/plugins/:id/test` | run `validate` → `{ ok }` or `{ ok:false, error }` |
+| `POST /api/plugins/:id/load` | enable + schedule + run now |
+| `POST /api/plugins/:id/unload` | unschedule + disable |
+| `POST /api/plugins/:id/run` | enqueue one immediate run |
+| `DELETE /api/plugins/:id` | unschedule + remove OpenBao secrets + delete row |
+| `GET /api/plugins/:id/runs` | `{ lastRunAt, lastStatus, lastError }` |
+
+## Scheduler internals
+
+The scheduler ([BullMQ](https://docs.bullmq.io/) over Redis) gives each instance
+a stable JobScheduler id (`plugin:<instanceId>`); load/unload upsert/remove
+that one schedule without disturbing the others. A daily `garbage_collect` job
+prunes discovery resources not seen in > 7 days.
+
+### Legacy migration
+
+Before this system, plugins were configured statically in `sso-secrets.js`:
+
+```javascript
+module.exports = {
+  discovery: {
+    plugins: {
+      proxmox: { enabled: true, cron: '0 * * * *', url: '…', tokenId: '…', tokenSecret: '…' }
+    }
+  }
+};
+```
+
+On the first boot of SSO Manager ≥ v1.17.0, if the `PluginInstance` table is
+empty **and** `conf.discovery.plugins` has entries, one instance per configured
+type is seeded automatically (secret fields copied into OpenBao). After that the
+table is non-empty and the static config is ignored — manage plugins from the
+UI/API instead. The migration is idempotent (guarded by the empty-table check).

docs/replication.md

@@ -0,0 +1,55 @@
+---
+layout: default
+title: Geo-Location Scaling (Replication)
+---
+
+# Geo-Location Scaling (Replication)
+
+SSO Manager is built to be a self-contained identity provider, but if you have multiple physical sites, you may want a local copy of the directory at each site to ensure low latency and high availability.
+
+## Why and when to use this?
+- **High Availability (HA)**: If your primary site goes completely offline, your other sites can still authenticate users locally without depending on a WAN link.
+- **Low Latency**: Applications at a remote site can bind directly to their local LDAP server (`localhost` or LAN IP) instead of traversing the internet to query the primary site, making logins blazing fast.
+- **Independent Failure Domains**: By replicating only the LDAP directory (the source of truth) and keeping session state (Redis) independent, you prevent complex "split-brain" scenarios in the web UI. A failure at Site A won't bring down Site B.
+
+By default, the `sso-manager` Docker container runs a single, independent OpenLDAP instance. However, you can enable **N-Way Multi-Master Replication** via environment variables.
+
+## How it works
+
+In an N-Way Multi-Master setup, every site runs a fully active OpenLDAP server (`slapd`).
+- **Reads and Writes anywhere**: A user can change their password or update their profile at Site A, Site B, or Site C.
+- **Conflict Resolution**: OpenLDAP's `syncrepl` engine uses Context Sequence Numbers (CSN) to track changes. If Site A goes offline and a user changes their password at Site B, Site A will automatically pull the newest changes the moment it rejoins the cluster.
+- **Independent Redis**: Session data, API Tokens, and OAuth Clients are stored in Redis. By design, Redis is NOT replicated in this geographic setup. This ensures that a failure at Site A never causes Site B's Redis to become read-only, which would break the web UI at Site B. OAuth clients must be configured per-site.
+
+## Configuration
+
+To enable replication, you must pass two environment variables to the `sso-manager` container:
+
+1. `LDAP_SERVER_ID`: A unique integer for this node (e.g., `1`, `2`, `3`). This MUST be unique across the cluster.
+2. `LDAP_REPLICATION_HOSTS`: A space-separated list of the LDAP URLs of all **other** nodes in the cluster.
+
+### Example using `theta-env` / Docker Compose
+
+**Site 1 (`setup.env` or `docker-compose.yml`)**
+```env
+LDAP_SERVER_ID=1
+LDAP_REPLICATION_HOSTS="ldaps://sso.site2.com:636 ldaps://sso.site3.com:636"
+```
+
+**Site 2 (`setup.env` or `docker-compose.yml`)**
+```env
+LDAP_SERVER_ID=2
+LDAP_REPLICATION_HOSTS="ldaps://sso.site1.com:636 ldaps://sso.site3.com:636"
+```
+
+**Site 3 (`setup.env` or `docker-compose.yml`)**
+```env
+LDAP_SERVER_ID=3
+LDAP_REPLICATION_HOSTS="ldaps://sso.site1.com:636 ldaps://sso.site2.com:636"
+```
+
+Once configured, the container's entrypoint will automatically load the `syncprov` module, enable `mirrormode`, and generate the necessary `syncrepl` blocks in `/etc/openldap/slapd.conf`.
+
+## User Locations
+
+When creating or editing a user, you can specify their **Location (Site)**. This maps directly to the standard LDAP `l` (localityName) attribute, allowing you to track which physical site a user belongs to natively within the directory.

docs/robots.txt

@@ -0,0 +1,4 @@
+User-agent: *
+Allow: /
+
+Sitemap: https://theta42.github.io/sso-manager-node/sitemap.xml

docs/site-join.md

@@ -0,0 +1,145 @@
+# Multi-Site: Joining a Spoke to the Master Directory
+
+The Directory can be deployed across multiple sites. The **master** site holds
+single write authority for the shared catalog; **spoke** sites run a read-only
+copy for local latency and autonomy (see the root `MULTI_SITE_SPEC.md` for the
+full architecture). This page covers the server endpoints that make a spoke
+"join" an existing master.
+
+> Status: **server endpoints + UI + setup.sh wiring, live replication, coordinated promotion.** A fresh bring-up can adopt a master directory via the Directory UI or via `setup.env`; a joined spoke is read-only with live WAN health, stays in sync after joining (not just a one-time snapshot), and can be promoted to master with the old master demoted as part of the same action.
+
+## The flow
+
+1. On the **master**, an admin mints a **site join key** (`stj_…`, shown once,
+   stored hashed, revocable) — Directory → the Master Site modal → **Site Join Keys**.
+2. On the **spoke** (a fresh install), either:
+   - **UI**: Directory → the Master Site modal → **Join an Existing Site**, or
+   - **setup.sh**: set `CFG_MASTER_DIRECTORY_URL` + `CFG_MASTER_DIRECTORY_JOIN_KEY`
+     in `setup.env` before the first run.
+3. The spoke pulls the master's directory export (LDAP tree + resource
+   catalog + agent-signing key), imports it, and persists its own spoke role
+   (`isMaster: false`, `masterUrl`, `siteSlug`) in `/config/site.json`.
+4. If the spoke also knows its own reachable URL (`selfUrl` — `setup.sh` passes
+   `https://$CFG_SSO_HOST` automatically), it registers itself with the master
+   (`POST /api/site/spokes`) so the master can push live updates back to it
+   afterward — see **Live replication** below. Without `selfUrl` the join still
+   succeeds; the spoke just stays a one-time snapshot.
+
+Joining is allowed only on a **fresh install** (no users beyond the bootstrap
+admin, no enrolled agents) — the join endpoint enforces this, so a populated
+directory can never be merged into a master's.
+
+## Live replication (not a one-time snapshot)
+
+A registered spoke stays in sync: every successful catalog write on the
+master fires a fire-and-forget push (`utils/site_replicate.js`) at every
+registered spoke, concurrently — one unreachable spoke never blocks or delays
+delivery to another. The spoke's `POST /api/site/resync` handler (called by
+that push) re-runs the same export-pull-and-import logic used at join time,
+so there's exactly one tested code path for "make my catalog match the
+master's," not a separate diff-application mechanism.
+
+The agent-signing key travels the same path: `POST /api/site/export`
+best-effort includes it, and the spoke adopts it via `agent_keys.adopt()` on
+both join and every resync. Every site holding the same signing key means any
+site's `sso-manager-node` can validly sign a command for any agent enrolled
+at any other site — a deliberate tradeoff (see `MULTI_SITE_SPEC.md` §2)
+accepted for this deployment's small, trusted scale. Don't extend this
+pattern to a larger/adversarial-tenant deployment without revisiting it.
+
+## Coordinated master promotion
+
+`POST /api/directory-admin/site-promote` (`god_admin` only) promotes this
+node to master as **one coordinated action**, not a manual two-step
+demote-then-promote:
+
+1. If this node currently has a master on file, it mints a fresh join key and
+   calls that master's `POST /api/site/demote` (authenticated with the join
+   key this node already holds), handing over the new key so the demoted node
+   can keep talking to the new master afterward.
+2. This step is **best-effort** — an unreachable old master (the WAN-outage
+   scenario this whole control exists for) never blocks the local promotion.
+   The response's `handoff` field reports what happened
+   (`"previous master demoted"`, an HTTP failure, or "unreachable, promoted
+   locally anyway") so the operator can reconcile it manually if needed.
+3. Every known spoke gets a fire-and-forget `master-promoted` resync ping so
+   they pick up the new master on their next sync.
+
+The Master Site modal's **Promote to Master** button surfaces the `handoff`
+result in a toast so the operator sees immediately whether the old master was
+actually reached.
+
+## Endpoints
+
+| Method | Path | Purpose |
+| :--- | :--- | :--- |
+| `GET` | `/api/site/join-keys` | List keys (prefix + usage only; never the key) |
+| `POST` | `/api/site/join-keys` | Mint one — returned **once** |
+| `POST` | `/api/site/join-keys/:id/revoke` | Stop it accepting new joins |
+| `DELETE` | `/api/site/join-keys/:id` | Remove it |
+| `GET` | `/api/site/config` | Current role (isMaster, masterUrl, siteSlug) |
+| `POST` | `/api/site/export` | Master directory export incl. agent-signing key (Bearer `stj_` key) |
+| `POST` | `/api/site/ping` | Lightweight master reachability probe (Bearer `stj_` key) |
+| `POST` | `/api/site/join` | Adopt a master directory + register for live replication (admin session) |
+| `POST` | `/api/site/spokes` | Register a spoke's endpoint for live replication (Bearer `stj_` key, called by the spoke right after join) |
+| `POST` | `/api/site/resync` | Re-pull the master's export (Bearer the spoke's own `pushToken`, called by the master's fire-and-forget push) |
+| `POST` | `/api/site/demote` | Step down to spoke of a new master (Bearer `stj_` key, called by the newly-promoted node) |
+| `POST` | `/api/directory-admin/site-promote` | Promote this node to master, coordinating demotion of the old one (`god_admin` session) |
+
+## Behavior after joining (spoke)
+
+- **Read-only**: directory-write requests (resources, edges, groups, secrets,
+  grants, driver actions, discovery merges) are rejected with `403` pointing at
+  the master. Writes must go to the master.
+- **WAN health**: `site-status` pings the master over the stored site join key
+  and reports `wanConnected`; the Master Site modal shows live Online/Offline.
+- **Role persists**: `isMaster`/`masterUrl`/`siteSlug` live in `/config/site.json`
+  (the env vars `IS_MASTER`/`MASTER_URL`/`SITE_SLUG` only seed the defaults), so
+  a restart never silently reverts a spoke to master.
+
+## Deployment (setup.sh)
+
+`setup.env` carries the intent so the join runs only on a **fresh** bring-up:
+
+```
+# Honored ONLY on first run; re-runs ignore it once ./config/ exists.
+CFG_MASTER_DIRECTORY_URL=https://sso.master.example.com
+CFG_MASTER_DIRECTORY_JOIN_KEY=stj_9f2e...
+```
+
+`setup.sh` runs `bootstrap/site-join.js` inside the sso-manager container after
+the bootstrap; it logs in as the admin and calls `/api/site/join`. A node that
+already joined reports "already a spoke" and setup continues (idempotent).
+
+## Security
+
+- Join keys are single-use-intent credentials: shown once, stored as a SHA-256
+  hash, revocable/expirable — the same model as agent join keys.
+- The export/ping endpoints return only the directory tree/catalog (no admin
+  secrets) and require a valid join key.
+- Join is admin-gated on the spoke, key-gated on the master, and fresh-install
+  gated on both sides.
+- The join key is stored on the spoke only so it can reach the master for WAN
+  health (and, in a later layer, write-proxy).
+- `pushToken` (the credential a spoke stores so it can recognize a legitimate
+  resync push from its master) is minted fresh per spoke registration and, by
+  design, kept in retrievable form on the master — unlike a join key, it's a
+  credential the master must keep *presenting*, not just verifying, so it
+  can't be one-way hashed. Compare `models/site_spoke.js`'s doc comment for
+  why that's the correct tradeoff, not an oversight.
+- Every site sharing one agent-signing key (see **Live replication** above)
+  means a compromised spoke — including the smallest, least-secured one — has
+  the same agent-command authority as the master. Accepted for this
+  deployment's scale; see `MULTI_SITE_SPEC.md` §2 before reusing this pattern
+  somewhere that assumption doesn't hold.
+
+## Not yet built
+
+- Traffic between sites (join/export/resync) still goes over the open
+  network path that already reaches the target — it does not route over the
+  WireGuard mesh `theta-gateway` can now establish (see `MULTI_SITE_SPEC.md`).
+- A no-inbound spoke (no public IP at all) still can't join — the mechanism
+  for a master to relay through the mesh to such a spoke is verified as
+  working, but nothing automates creating that route yet.
+- OpenBao secret replication covers only the agent-signing key; LDAP admin
+  creds, JWT secret, and other per-deployment secrets aren't synced.

docs/vault.md

@@ -0,0 +1,58 @@
+---
+layout: default
+title: Vault Secrets
+description: OpenBao-backed personal, shared, and external-app secret storage built into the SSO Manager UI.
+---
+
+# Vault Secrets Management
+
+[← Back to Home](index.html)
+
+The Vault Secrets feature integrates with OpenBao to provide a secure key-value store for your environment. It allows you to store sensitive information like passwords, API keys, and credentials, ensuring they are encrypted and access-controlled.
+
+## Location & Access
+
+- **External App Tokens**: Managed under **Configuration** (`/conf` -> **App Tokens** tab). Admins can mint and view periodic OpenBao app tokens scoped to `secret/apps/<name>/*`.
+- **Resource Secrets**: Managed under **Directory** (`/directory`) inside each resource's modal under the **Secrets** tab. Stored in OpenBao under `secret/data/resources/<slug>/conf`.
+
+## External App Tokens (Admin)
+
+The **App Tokens** tab in **Configuration** (`/conf`) mints a scoped OpenBao token for an **external application** or script so it can read its own configuration out of OpenBao.
+
+1. Enter an app **name** (e.g. `build-agent`) and click **Mint token**.
+2. A token is shown **once** — copy it into the external app now; it cannot be recovered later. The app uses it as the `X-Vault-Token` header against `secret/apps/<name>/*`.
+3. The **Active App Tokens** list shows every token created (metadata only — the token itself is never stored). sso-manager keeps each token alive by renewing it periodically.
+
+The token is strictly scoped to `secret/apps/<name>/*` (policy `app-<name>`), so a compromised token can't touch any other secret.
+
+## Shared tab
+
+The **Shared** tab lets you share a secret with another user (or app) without copying the value around.
+
+1. **New** — give the secret a name (slug) and its JSON data. The owner has full read/write on `secret/shared/<uid>/<slug>`.
+2. Open a secret and use **Grants** to share it with a user or app; the grantee's OpenBao policy is edited immediately so the share takes effect with no token re-mint. Revoking a grant removes access at the ACL.
+3. The data itself is read through the normal Vault proxy using each user's own session, so OpenBao enforces read access per-request.
+
+## API Access
+
+To read your own secrets programmatically, call the `/api/vault` proxy with
+a [personal API token](concepts-api-tokens.html) — **not** a raw OpenBao
+token. The server authenticates the request, resolves your own scoped
+OpenBao access, and injects the real `X-Vault-Token` itself:
+
+```bash
+# Example: Read a secret via the API (KV-v2, so the path includes /data/)
+curl -H "Authorization: Bearer sso_<id>_<secret>" \
+  https://<your-sso-host>/api/vault/secret/data/<your-secret-path>
+```
+
+An **external app** reading its own config uses the scoped token minted for
+it on the **Apps** tab instead of a personal token — see *Apps tab (admin)*
+above for how that token is minted and what it's confined to.
+
+Using the OpenBao **root token** directly (bypassing the SSO entirely) is
+never the intended path for day-to-day secret access — it's an
+operator/maintenance credential (seeding, disaster recovery), kept in
+`setup.env` and never passed to a service container. See
+[theta-env's Secrets doc](https://theta42.github.io/theta-env/secrets.html)
+for the full token/policy model.

install.sh

@@ -1,719 +1,230 @@
 #!/usr/bin/env bash
-# install.sh - Idempotent standalone installer for Theta42 SSO Manager
-# For Debian/Ubuntu systems
 #
-# This script:
-#   1. Installs Node.js 20.x
-#   2. Installs and configures OpenLDAP with required schemas/overlays
-#   3. Deploys the SSO Manager application
-#   4. Sets up systemd services
+# Install / update Theta42 SSO Manager on a fresh or existing host.
 #
-# Usage:
-#   sudo ./install.sh [OPTIONS]
+# This script is idempotent: run it to install, and re-run it to update. It
+# installs system dependencies (Node, OpenLDAP, Redis), force-syncs the repo at
+# $REPO_DIR to its remote branch, and symlinks the systemd config straight from
+# the repo. Because the config is symlinked, an update is just "sync the repo +
+# restart" -- the files under /etc/systemd always track the repo.
 #
-# Options:
-#   -p, --admin-pass PASSWORD    LDAP admin password (required, or set via LDAP_ADMIN_PASS env)
-#   -b, --base-dn DN             Base DN (default: dc=example,dc=com)
-#   -n, --org-name NAME          Organization name shown in UI/email (default: SSO Manager)
-#   -o, --port PORT              HTTP port for SSO Manager (default: 3001)
-#   -j, --jwt-secret SECRET      JWT secret for OAuth (default: auto-generated)
-#   -s, --smtp-config CONFIG     SMTP config as host:port:user:pass
-#   --skip-ldap                  Skip LDAP installation (use existing LDAP)
-#   --skip-app                   Skip application installation (LDAP setup only)
-#   --dry-run                    Show what would be done without making changes
-#   -h, --help                   Show this help
+# Secrets live at $SECRETS_FILE (/etc/sso-manager/secrets.js by default),
+# outside the repo checkout so they survive the hard reset below. FIRST RUN
+# ONLY (no $SECRETS_FILE yet): installs and configures OpenLDAP (modules,
+# overlays, custom schema, directory tree, required SSO groups -- see
+# ops/ldap-setup.sh), generates an LDAP admin password + JWT secret unless
+# given via env, and seeds $SECRETS_FILE with those values plus SMTP
+# placeholders. Edit that file (SMTP, org name, ...) and re-run this script to
+# apply changes -- once it exists it is never touched again, and LDAP is never
+# re-bootstrapped.
 #
-# Environment variables (alternative to flags):
-#   LDAP_ADMIN_PASS, LDAP_BASE_DN, PORT, JWT_SECRET, SMTP_*
-
+# Intended to be driven by CI/CD with no human writes on prod: the checkout is
+# hard-reset to origin/$BRANCH on every run, so the box deterministically
+# mirrors the repo (any drift on the box is discarded).
+#
+# Usage: sudo ./install.sh   (override with REPO_URL=, REPO_DIR=, BRANCH=,
+#                             SECRETS_FILE=, LDAP_BASE_DN=, LDAP_ADMIN_PASS=,
+#                             JWT_SECRET=, ORG_NAME=, PORT=, SKIP_LDAP=true)
 set -euo pipefail
+# Never block on an interactive git credential prompt in CI.
+export GIT_TERMINAL_PROMPT=0
+# Never block on an interactive debconf prompt (e.g. tzdata, pulled in as a
+# dependency of redis-server/slapd on a box that's never configured it).
+export DEBIAN_FRONTEND=noninteractive
 
-# ── Defaults ──────────────────────────────────────────────────────────────────
-BASE_DN="${LDAP_BASE_DN:-dc=example,dc=com}"
-ADMIN_PASS="${LDAP_ADMIN_PASS:-}"
+REPO_URL="${REPO_URL:-https://github.com/theta42/sso-manager-node.git}"
+REPO_DIR="${REPO_DIR:-/opt/theta42/sso-manager}"
+BRANCH="${BRANCH:-master}"
+NODE_MAJOR=22
+SECRETS_FILE="${SECRETS_FILE:-/etc/sso-manager/secrets.js}"
+
+LDAP_BASE_DN="${LDAP_BASE_DN:-dc=example,dc=com}"
 ORG_NAME="${ORG_NAME:-SSO Manager}"
 PORT="${PORT:-3001}"
-JWT_SECRET="${JWT_SECRET:-}"
-SMTP_HOST="${SMTP_HOST:-}"
-SMTP_PORT="${SMTP_PORT:-587}"
-SMTP_USER="${SMTP_USER:-}"
-SMTP_PASS="${SMTP_PASS:-}"
 SKIP_LDAP="${SKIP_LDAP:-false}"
-SKIP_APP="${SKIP_APP:-false}"
-DRY_RUN="${DRY_RUN:-false}"
 
-INSTALL_DIR="/opt/sso-manager"
-SYSTEMD_DIR="/etc/systemd/system"
-SCRIPT_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)"
-
-# Colors for output
-RED='\033[0;31m'
-GREEN='\033[0;32m'
-YELLOW='\033[1;33m'
-NC='\033[0m' # No Color
-
-# ── Helper functions ──────────────────────────────────────────────────────────
-info() { echo -e "${GREEN}[INFO]${NC} $*"; }
-warn() { echo -e "${YELLOW}[WARN]${NC} $*" >&2; }
-error() { echo -e "${RED}[ERROR]${NC} $*" >&2; }
-dry_run() { if [[ "$DRY_RUN" == "true" ]]; then echo "[DRY-RUN] $*"; fi; }
-
-usage() {
-    grep '^#' "$0" | sed 's/^# \{0,1\}//'
-    exit 0
-}
-
-# Parse arguments
-while [[ $# -gt 0 ]]; do
-    case $1 in
-        -p|--admin-pass)
-            ADMIN_PASS="$2"
-            shift 2
-            ;;
-        -b|--base-dn)
-            BASE_DN="$2"
-            shift 2
-            ;;
-        -n|--org-name)
-            ORG_NAME="$2"
-            shift 2
-            ;;
-        -o|--port)
-            PORT="$2"
-            shift 2
-            ;;
-        -j|--jwt-secret)
-            JWT_SECRET="$2"
-            shift 2
-            ;;
-        -s|--smtp-config)
-            IFS=':' read -r SMTP_HOST SMTP_PORT SMTP_USER SMTP_PASS <<< "$2"
-            shift 2
-            ;;
-        --skip-ldap)
-            SKIP_LDAP="true"
-            shift
-            ;;
-        --skip-app)
-            SKIP_APP="true"
-            shift
-            ;;
-        --dry-run)
-            DRY_RUN="true"
-            shift
-            ;;
-        -h|--help)
-            usage
-            ;;
-        *)
-            error "Unknown option: $1"
-            usage
-            ;;
-    esac
-done
-
-# Validate required parameters
-if [[ -z "$ADMIN_PASS" ]]; then
-    error "LDAP admin password is required (-p or LDAP_ADMIN_PASS env)"
-    exit 1
+if [ "$(id -u)" -ne 0 ]; then
+	echo "This script must be run as root (try: sudo $0)" >&2
+	exit 1
 fi
 
-# Generate JWT secret if not provided
-if [[ -z "$JWT_SECRET" ]]; then
-    JWT_SECRET=$(openssl rand -hex 32)
-    info "Generated JWT secret: ${JWT_SECRET:0:8}..."
+# Symlink $1 -> $2, replacing whatever is already at $2 (idempotent).
+link(){
+	ln -sfn "$1" "$2"
+	echo "linked $2 -> $1"
+}
+
+# Read the "version" field out of a package.json without depending on Node
+# being installed yet (this runs before the Node.js install step below).
+pkg_version(){
+	sed -n 's/^[[:space:]]*"version":[[:space:]]*"\([^"]*\)".*/\1/p' "$1" | head -1
+}
+
+# Installed version before this run touches anything, for the upgrade banner
+# at the end. Empty on a fresh install (no prior checkout).
+CURRENT_VERSION=""
+if [ -f "$REPO_DIR/nodejs/package.json" ]; then
+	CURRENT_VERSION="$(pkg_version "$REPO_DIR/nodejs/package.json")"
 fi
 
-# Derive the DNS domain from the base DN (dc=foo,dc=bar -> foo.bar) for email
-# sender defaults. Override with LDAP_DOMAIN if set.
-if [[ -z "${LDAP_DOMAIN:-}" ]]; then
-    LDAP_DOMAIN=$(echo "$BASE_DN" | sed 's/^dc=//; s/,dc=/./g')
+# FIRST_RUN gates OpenLDAP bootstrap + secrets seeding below -- both only ever
+# happen once, the first time this script runs on a host (i.e. before
+# $SECRETS_FILE exists). Every later run only updates the code.
+FIRST_RUN=0
+[ -f "$SECRETS_FILE" ] || FIRST_RUN=1
+
+echo "==> Base packages"
+apt-get update
+apt-get install -y --no-install-recommends \
+	build-essential redis-server \
+	wget gnupg ca-certificates curl git
+
+echo "==> Node.js ${NODE_MAJOR}.x apt source"
+install -d -m 0755 /etc/apt/keyrings
+curl -fsSL https://deb.nodesource.com/gpgkey/nodesource-repo.gpg.key \
+	| gpg --dearmor --yes -o /etc/apt/keyrings/nodesource.gpg
+echo "deb [signed-by=/etc/apt/keyrings/nodesource.gpg] https://deb.nodesource.com/node_${NODE_MAJOR}.x nodistro main" \
+	> /etc/apt/sources.list.d/nodesource.list
+
+echo "==> Install Node.js"
+apt-get update
+apt-get install -y nodejs
+
+echo "==> Redis"
+systemctl enable --now redis-server
+
+echo "==> Repo checkout at ${REPO_DIR} (branch ${BRANCH})"
+install -d "$(dirname "$REPO_DIR")"
+if [ -d "$REPO_DIR/.git" ]; then
+	# Force the box to match the remote branch exactly. No human edits configs
+	# on prod, so discarding local drift is the desired, deterministic behavior.
+	git -C "$REPO_DIR" fetch --prune origin
+	git -C "$REPO_DIR" checkout -B "$BRANCH" "origin/$BRANCH"
+	git -C "$REPO_DIR" reset --hard "origin/$BRANCH"
+	git -C "$REPO_DIR" clean -fd
+else
+	git clone --branch "$BRANCH" "$REPO_URL" "$REPO_DIR"
 fi
 
-# ── System checks ─────────────────────────────────────────────────────────────
-check_root() {
-    if [[ $EUID -ne 0 ]]; then
-        error "This script must be run as root (sudo)"
-        exit 1
-    fi
-}
-
-check_os() {
-    if [[ ! -f /etc/debian_version ]]; then
-        error "This script is for Debian/Ubuntu systems only"
-        exit 1
-    fi
-    info "Detected $(cat /etc/os-release | grep PRETTY_NAME | cut -d'"' -f2)"
-}
-
-# ── Package installation ──────────────────────────────────────────────────────
-install_package() {
-    local pkg="$1"
-    if dpkg -l | grep -q "^ii  $pkg "; then
-        info "Package $pkg is already installed"
-        return 0
-    fi
-    dry_run "Would install package: $pkg"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-    apt-get update -qq
-    apt-get install -y -qq "$pkg"
-    info "Installed $pkg"
-}
-
-install_nodejs() {
-    if command -v node &>/dev/null && node --version | grep -q "v20"; then
-        info "Node.js 20.x is already installed"
-        return 0
-    fi
-    dry_run "Would install Node.js 20.x"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-
-    info "Installing Node.js 20.x..."
-    # Use NodeSource repository for Node.js 20.x
-    apt-get update -qq
-    apt-get install -y -qq curl gnupg ca-certificates
-    curl -fsSL https://deb.nodesource.com/setup_20.x | bash - >/dev/null 2>&1
-    apt-get install -y -qq nodejs
-    info "Installed Node.js $(node --version)"
-}
-
-# ── OpenLDAP installation and configuration ───────────────────────────────────
-install_openldap() {
-    if command -v slapd &>/dev/null; then
-        info "OpenLDAP is already installed"
-        return 0
-    fi
-    dry_run "Would install OpenLDAP"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-
-    info "Installing OpenLDAP..."
-
-    # Pre-seed debconf for non-interactive installation
-    debconf-set-selections << EOF
-slapd slapd/internal/adminpw string $ADMIN_PASS
-slapd slapd/password1 string $ADMIN_PASS
-slapd slapd/password2 string $ADMIN_PASS
-slapd slapd/domain string ${BASE_DN#dc=}
-slapd slapd/backend string MDB
-slapd shared/organization string $ORG_NAME
-slapd slapd/purge_database boolean true
-slapd slapd/move_old_database boolean true
-slapd slapd/invalid_config boolean true
-EOF
-
-    apt-get update -qq
-    apt-get install -y -qq slapd ldap-utils
-
-    # Configure ldap.conf
-    cat > /etc/ldap/ldap.conf << LDAPCONF
-BASE   $BASE_DN
-URI    ldap://localhost
-LDAPCONF
-
-    # Set proper permissions
-    chmod 644 /etc/ldap/ldap.conf
-
-    info "OpenLDAP installed"
-}
-
-configure_openldap() {
-    info "Configuring OpenLDAP..."
-    dry_run "Would configure OpenLDAP with base DN: $BASE_DN"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-
-    # Wait for slapd to be ready
-    for i in {1..10}; do
-        if ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" "(objectClass=*)" dn >/dev/null 2>&1; then
-            info "OpenLDAP is ready"
-            break
-        fi
-        sleep 1
-    done
-
-    # Detect the database DN for our suffix
-    DB_DN=$(ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" \
-        "(&(objectClass=olcDatabaseConfig)(olcSuffix=${BASE_DN}))" dn 2>/dev/null \
-        | grep "^dn:" | head -1 | sed 's/^dn: //')
-
-    if [[ -z "$DB_DN" ]]; then
-        # Try to find any database and update its suffix
-        DB_DN=$(ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" \
-            "(objectClass=olcDatabaseConfig)" dn 2>/dev/null \
-            | grep "^dn:" | head -1 | sed 's/^dn: //')
-
-        if [[ -n "$DB_DN" ]]; then
-            info "Updating database suffix to $BASE_DN"
-            ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: $DB_DN
-changetype: modify
-replace: olcSuffix
-olcSuffix: $BASE_DN
-EOF
-        fi
-    fi
-
-    if [[ -z "$DB_DN" ]]; then
-        error "Could not detect OpenLDAP database configuration"
-        return 1
-    fi
-
-    info "Using database: $DB_DN"
-
-    # 1. Load pw-sha2 module
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" "(objectClass=olcModuleList)" olcModuleLoad 2>/dev/null | grep -q "pw-sha2"; then
-        info "Loading pw-sha2 module..."
-        ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: cn=module{0},cn=config
-changetype: modify
-add: olcModuleLoad
-olcModuleLoad: pw-sha2
-EOF
-    else
-        info "pw-sha2 module already loaded"
-    fi
-
-    # 2. Load ppolicy module
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" "(objectClass=olcModuleList)" olcModuleLoad 2>/dev/null | grep -q "ppolicy"; then
-        info "Loading ppolicy module..."
-        ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: cn=module{0},cn=config
-changetype: modify
-add: olcModuleLoad
-olcModuleLoad: ppolicy
-EOF
-    else
-        info "ppolicy module already loaded"
-    fi
-
-    # 3. Load memberof module
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" "(objectClass=olcModuleList)" olcModuleLoad 2>/dev/null | grep -q "memberof"; then
-        info "Loading memberof module..."
-        ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: cn=module{1},cn=config
-changetype: modify
-add: olcModuleLoad
-olcModuleLoad: memberof
-EOF
-    else
-        info "memberof module already loaded"
-    fi
-
-    # 4. Load refint module
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=config" "(objectClass=olcModuleList)" olcModuleLoad 2>/dev/null | grep -q "refint"; then
-        info "Loading refint module..."
-        ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: cn=module{1},cn=config
-changetype: modify
-add: olcModuleLoad
-olcModuleLoad: refint
-EOF
-    else
-        info "refint module already loaded"
-    fi
-
-    # 5. Add ppolicy overlay
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "$DB_DN" "(olcOverlay=*ppolicy*)" dn 2>/dev/null | grep -qi "ppolicy"; then
-        info "Adding ppolicy overlay..."
-        ldapadd -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: olcOverlay=ppolicy,$DB_DN
-objectClass: olcOverlayConfig
-objectClass: olcPPolicyConfig
-olcOverlay: ppolicy
-olcPPolicyDefault: cn=ppolicy,ou=policies,$BASE_DN
-olcPPolicyUseLockout: TRUE
-EOF
-    else
-        info "ppolicy overlay already configured"
-    fi
-
-    # 6. Add memberof overlay
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "$DB_DN" "(olcOverlay=*memberof*)" dn 2>/dev/null | grep -qi "memberof"; then
-        info "Adding memberof overlay..."
-        ldapadd -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: olcOverlay=memberof,$DB_DN
-objectClass: olcConfig
-objectClass: olcMemberOf
-objectClass: olcOverlayConfig
-objectClass: top
-olcOverlay: memberof
-olcMemberOfDangling: ignore
-olcMemberOfRefInt: TRUE
-olcMemberOfGroupOC: groupOfNames
-olcMemberOfMemberAD: member
-olcMemberOfMemberOfAD: memberOf
-EOF
-    else
-        info "memberof overlay already configured"
-    fi
-
-    # 7. Add refint overlay
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "$DB_DN" "(olcOverlay=*refint*)" dn 2>/dev/null | grep -qi "refint"; then
-        info "Adding refint overlay..."
-        ldapadd -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: olcOverlay=refint,$DB_DN
-objectClass: olcConfig
-objectClass: olcOverlayConfig
-objectClass: olcRefintConfig
-objectClass: top
-olcOverlay: refint
-olcRefintAttribute: memberof member manager owner
-EOF
-    else
-        info "refint overlay already configured"
-    fi
-
-    # 8. Add database indexes
-    info "Configuring database indexes..."
-    for index in "mail eq,sub" "uid eq,sub" "cn eq,sub" "member eq" "uidNumber eq" "gidNumber eq"; do
-        attr=$(echo "$index" | cut -d' ' -f1)
-        types=$(echo "$index" | cut -d' ' -f2)
-        ldapmodify -Q -Y EXTERNAL -H ldapi:/// << EOF || true
-dn: $DB_DN
-changetype: modify
-add: olcDbIndex
-olcDbIndex: $attr $types
-EOF
-    done
-
-    # 9. Load custom theta42 schema
-    if ! ldapsearch -Q -Y EXTERNAL -H ldapi:/// -b "cn=schema,cn=config" "(olcObjectClasses=*theta42Person*)" olcObjectClasses 2>/dev/null | grep -q "theta42"; then
-        info "Loading custom theta42 schema..."
-        ldapadd -Q -Y EXTERNAL -H ldapi:/// << EOF
-dn: cn=theta42,cn=schema,cn=config
-objectClass: olcSchemaConfig
-cn: theta42
-olcAttributeTypes: ( 1.3.6.1.4.1.99999.1.1
-  NAME 'dateOfBirth'
-  DESC 'Date of birth in ISO 8601 format YYYY-MM-DD'
-  EQUALITY caseExactMatch
-  SUBSTR caseExactSubstringsMatch
-  SYNTAX 1.3.6.1.4.1.1466.115.121.1.15
-  SINGLE-VALUE )
-olcObjectClasses: ( 1.3.6.1.4.1.99999.2.1
-  NAME 'theta42Person'
-  DESC 'Theta42 SSO extended person attributes'
-  AUXILIARY
-  MAY ( dateOfBirth ) )
-EOF
-    else
-        info "theta42 schema already loaded"
-    fi
-
-    # 10. Create base directory structure
-    BIND_DN="cn=admin,$BASE_DN"
-
-    # Create base DN if it doesn't exist
-    if ! ldapsearch -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// -b "$BASE_DN" -s base "(objectClass=*)" dn 2>/dev/null | grep -q "dn:"; then
-        info "Creating base DN structure..."
-        DC_VALUE="${BASE_DN#dc=}"
-        DC_VALUE="${DC_VALUE%%,*}"
-
-        ldapadd -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// << EOF
-dn: $BASE_DN
-objectClass: dcObject
-objectClass: organization
-dc: $DC_VALUE
-o: $ORG_NAME
-EOF
-    else
-        info "Base DN already exists"
-    fi
-
-    # Create OUs
-    for ou in people groups policies; do
-        dn="ou=$ou,$BASE_DN"
-        if ! ldapsearch -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// -b "$dn" -s base "(objectClass=*)" dn 2>/dev/null | grep -q "dn:"; then
-            info "Creating $ou OU..."
-            ldapadd -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// << EOF
-dn: ou=$ou,$BASE_DN
-objectClass: organizationalUnit
-ou: $ou
-EOF
-        else
-            info "OU $ou already exists"
-        fi
-    done
-
-    # 11. Create default ppolicy
-    if ! ldapsearch -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// -b "cn=ppolicy,ou=policies,$BASE_DN" -s base "(objectClass=*)" dn 2>/dev/null | grep -q "dn:"; then
-        info "Creating default ppolicy..."
-        ldapadd -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// << EOF
-dn: cn=ppolicy,ou=policies,$BASE_DN
-objectClass: top
-objectClass: organizationalRole
-objectClass: pwdPolicy
-cn: ppolicy
-pwdAttribute: 2.5.4.35
-pwdLockout: FALSE
-pwdMustChange: FALSE
-pwdAllowUserChange: TRUE
-EOF
-    else
-        info "Default ppolicy already exists"
-    fi
-
-    # 12. Create required SSO groups
-    for group in app_sso_admin app_sso_invite app_sso_oauth_admin; do
-        dn="cn=$group,ou=groups,$BASE_DN"
-        if ! ldapsearch -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// -b "$dn" -s base "(objectClass=*)" dn 2>/dev/null | grep -q "dn:"; then
-            info "Creating group: $group"
-            ldapadd -x -D "$BIND_DN" -w "$ADMIN_PASS" -H ldapi:/// << EOF
-dn: $dn
-objectClass: groupOfNames
-objectClass: top
-cn: $group
-description: $ORG_NAME $group group
-member: $BIND_DN
-EOF
-        else
-            info "Group $group already exists"
-        fi
-    done
-
-    info "OpenLDAP configuration complete"
-}
-
-# ── Application installation ──────────────────────────────────────────────────
-install_app() {
-    info "Installing SSO Manager application..."
-    dry_run "Would install application to $INSTALL_DIR"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-
-    # Create installation directory
-    mkdir -p "$INSTALL_DIR"
-
-    # Copy application files
-    info "Copying application files..."
-    cp -r "$SCRIPT_DIR/nodejs/"* "$INSTALL_DIR/"
-
-    # Install npm dependencies
-    info "Installing npm dependencies..."
-    cd "$INSTALL_DIR"
-    npm ci --only=production --quiet
-
-    # Create secrets configuration
-    info "Creating application configuration..."
-    cat > "$INSTALL_DIR/conf/secrets.js" << SECRETEOF
-'use strict';
-
-module.exports = {
-    port: $PORT,
-    ldap: {
-        url: 'ldap://localhost',
-        bindDN: 'cn=admin,$BASE_DN',
-        bindPassword: '$ADMIN_PASS',
-        userBase: 'ou=people,$BASE_DN',
-        groupBase: 'ou=groups,$BASE_DN',
-    },
-    smtp: {
-        host: '${SMTP_HOST:-localhost}',
-        port: ${SMTP_PORT:-587},
-        user: '${SMTP_USER:-}',
-        pass: '${SMTP_PASS:-}',
-        from: '${ORG_NAME} <noreply@${LDAP_DOMAIN}>',
-    },
-    voipms: {
-        username: '${VOIPMS_USER:-}',
-        password: '${VOIPMS_PASS:-}',
-        did: '${VOIPMS_DID:-}',
-    },
-    oauth: {
-        issuer: '',
-        jwtSecret: '$JWT_SECRET',
-        token_lifetime: {
-            access_token: 3600,
-            refresh_token: 2592000
-        }
-    },
-};
-SECRETEOF
-
-    # Create base configuration
-    cat > "$INSTALL_DIR/conf/base.js" << BASEEOF
-'use strict';
-
-module.exports = {
-    name: "$ORG_NAME",
-    userModel: 'ldap',
-    redis: {
-        prefix: 'sso_manager_'
-    },
-    ldap: {
-        url: 'ldap://localhost',
-        bindDN: 'cn=admin,$BASE_DN',
-        bindPassword: '__IN SECRETS FILE__',
-        userBase: 'ou=people,$BASE_DN',
-        groupBase: 'ou=groups,$BASE_DN',
-        userFilter: '(objectClass=posixAccount)',
-        userNameAttribute: 'uid'
-    },
-    oauth: {
-        issuer: '',
-        jwtSecret: '__in secrets file__',
-        token_lifetime: {
-            access_token: 3600,
-            refresh_token: 2592000
-        }
-    },
-    smtp: {
-        host: 'localhost',
-        port: 587,
-        secure: false,
-        from: '$ORG_NAME <noreply@$LDAP_DOMAIN>',
-    },
-};
-BASEEOF
-
-    # Set ownership
-    chown -R root:root "$INSTALL_DIR"
-    chmod -R 755 "$INSTALL_DIR"
-
-    info "Application installed to $INSTALL_DIR"
-}
-
-# ── Systemd service configuration ─────────────────────────────────────────────
-install_systemd() {
-    info "Installing systemd service..."
-    dry_run "Would install systemd service"
-    [[ "$DRY_RUN" == "true" ]] && return 0
-
-    cat > "$SYSTEMD_DIR/sso-manager.service" << UNITEOF
-[Unit]
-Description=Theta42 SSO Manager
-Documentation=file://$INSTALL_DIR/README.md
-After=network.target slapd.service
-Wants=slapd.service
-
-[Service]
-Type=simple
-User=root
-WorkingDirectory=$INSTALL_DIR
-ExecStart=/usr/bin/node $INSTALL_DIR/bin/www
-Restart=on-failure
-RestartSec=5
-Environment=NODE_ENV=production
-Environment=NODE_PORT=$PORT
-
-# Security hardening
-NoNewPrivileges=true
-PrivateTmp=true
-
-[Install]
-WantedBy=multi-user.target
-UNITEOF
-
-    systemctl daemon-reload
-    systemctl enable sso-manager.service
-
-    info "Systemd service installed"
-}
-
-# ── Verification ──────────────────────────────────────────────────────────────
-verify_installation() {
-    info "Verifying installation..."
-
-    local errors=0
-
-    # Check OpenLDAP
-    if command -v slapd &>/dev/null; then
-        if systemctl is-active --quiet slapd; then
-            info "✓ OpenLDAP is running"
-        else
-            warn "✗ OpenLDAP is not running"
-            ((errors++))
-        fi
-    else
-        warn "✗ OpenLDAP is not installed"
-        ((errors++))
-    fi
-
-    # Check application
-    if [[ -d "$INSTALL_DIR" ]]; then
-        info "✓ Application is installed"
-    else
-        warn "✗ Application is not installed"
-        ((errors++))
-    fi
-
-    # Check systemd service
-    if systemctl is-enabled --quiet sso-manager.service 2>/dev/null; then
-        info "✓ Systemd service is enabled"
-    else
-        warn "✗ Systemd service is not enabled"
-        ((errors++))
-    fi
-
-    if [[ $errors -eq 0 ]]; then
-        info "Installation verified successfully"
-    else
-        warn "Installation completed with $errors issue(s)"
-    fi
-
-    return $errors
-}
-
-# ── Main execution ────────────────────────────────────────────────────────────
-main() {
-    echo
-    echo "=============================================="
-    echo "  Theta42 SSO Manager Installer"
-    echo "=============================================="
-    echo
-    echo "Configuration:"
-    echo "  Base DN:     $BASE_DN"
-    echo "  Port:        $PORT"
-    echo "  Install dir: $INSTALL_DIR"
-    echo "  Skip LDAP:   $SKIP_LDAP"
-    echo "  Skip App:    $SKIP_APP"
-    echo
-
-    check_root
-    check_os
-
-    if [[ "$SKIP_LDAP" != "true" ]]; then
-        echo
-        info "=== Installing OpenLDAP ==="
-        install_openldap
-        configure_openldap
-    fi
-
-    if [[ "$SKIP_APP" != "true" ]]; then
-        echo
-        info "=== Installing SSO Manager ==="
-        install_nodejs
-        install_app
-        install_systemd
-    fi
-
-    echo
-    verify_installation
-
-    echo
-    echo "=============================================="
-    echo "  Installation Complete!"
-    echo "=============================================="
-    echo
-
-    if [[ "$SKIP_APP" != "true" ]]; then
-        info "Start the service with: systemctl start sso-manager"
-        info "View logs with: journalctl -fu sso-manager"
-        info "Access the UI at: http://localhost:$PORT"
-    fi
-
-    if [[ "$SKIP_LDAP" != "true" ]]; then
-        echo
-        info "LDAP Configuration:"
-        info "  Base DN:     $BASE_DN"
-        info "  Bind DN:     cn=admin,$BASE_DN"
-        info "  Admin pass:  (set by you)"
-        echo
-        info "Required SSO groups created:"
-        info "  - app_sso_admin"
-        info "  - app_sso_invite"
-        info "  - app_sso_oauth_admin"
-    fi
-
-    echo
-}
-
-main
+NEW_VERSION="$(pkg_version "$REPO_DIR/nodejs/package.json")"
+
+if [ "$FIRST_RUN" -eq 1 ] && [ "$SKIP_LDAP" != "true" ]; then
+	echo "==> First run: bootstrapping OpenLDAP (base DN: ${LDAP_BASE_DN})"
+	LDAP_ADMIN_PASS="${LDAP_ADMIN_PASS:-$(openssl rand -base64 24 | tr -d '=+/')}"
+	JWT_SECRET="${JWT_SECRET:-$(openssl rand -hex 32)}"
+	BIND_DN="cn=admin,${LDAP_BASE_DN}"
+	# slapd/domain wants a dotted DNS domain (e.g. "example.com"), not the raw
+	# DN -- "dc=foo,dc=bar" -> "foo.bar". A malformed value here (e.g. the raw
+	# DN with only the leading "dc=" stripped) makes slapd's postinst hang
+	# indefinitely instead of failing cleanly.
+	LDAP_DOMAIN="$(echo "$LDAP_BASE_DN" | sed 's/^dc=//; s/,dc=/./g')"
+
+	if ! command -v slapd >/dev/null 2>&1; then
+		debconf-set-selections <<-EOF
+		slapd slapd/internal/adminpw password ${LDAP_ADMIN_PASS}
+		slapd slapd/password1 password ${LDAP_ADMIN_PASS}
+		slapd slapd/password2 password ${LDAP_ADMIN_PASS}
+		slapd slapd/domain string ${LDAP_DOMAIN}
+		slapd shared/organization string ${ORG_NAME}
+		slapd slapd/purge_database boolean true
+		slapd slapd/move_old_database boolean true
+		EOF
+		apt-get install -y slapd ldap-utils
+		cat > /etc/ldap/ldap.conf <<-EOF
+		BASE   ${LDAP_BASE_DN}
+		URI    ldap://localhost
+		EOF
+		systemctl enable --now slapd
+	else
+		echo "    slapd already installed -- assuming it already serves ${LDAP_BASE_DN}"
+	fi
+
+	echo "==> Directory structure (ou=people, ou=groups)"
+	for ou in people groups; do
+		dn="ou=${ou},${LDAP_BASE_DN}"
+		if ldapsearch -x -D "$BIND_DN" -w "$LDAP_ADMIN_PASS" -H ldap://localhost -b "$dn" -s base "(objectClass=*)" dn 2>/dev/null | grep -q "^dn:"; then
+			echo "    ${dn} already exists"
+		else
+			ldapadd -x -D "$BIND_DN" -w "$LDAP_ADMIN_PASS" -H ldap://localhost <<-EOF
+			dn: ${dn}
+			objectClass: organizationalUnit
+			ou: ${ou}
+			EOF
+			echo "    ${dn} created"
+		fi
+	done
+
+	echo "==> LDAP modules, overlays, schema, policy, SSO groups"
+	"$REPO_DIR/ops/ldap-setup.sh" -p "$LDAP_ADMIN_PASS" -b "$LDAP_BASE_DN" -D "$BIND_DN"
+
+	echo "==> Seeding ${SECRETS_FILE}"
+	install -d -m 0750 "$(dirname "$SECRETS_FILE")"
+	cat > "$SECRETS_FILE" <<-SECRETSEOF
+	'use strict';
+
+	// Generated by install.sh on $(date -u +%Y-%m-%dT%H:%M:%SZ). Edit freely --
+	// this file is never overwritten by a later run of install.sh.
+	// LDAP admin password + JWT secret below were auto-generated; SMTP is a
+	// placeholder (email delivery won't work until you fill it in).
+
+	module.exports = {
+	    port: ${PORT},
+	    name: '${ORG_NAME}',
+	    ldap: {
+	        url: 'ldap://localhost',
+	        bindDN: '${BIND_DN}',
+	        bindPassword: '${LDAP_ADMIN_PASS}',
+	        userBase: 'ou=people,${LDAP_BASE_DN}',
+	        groupBase: 'ou=groups,${LDAP_BASE_DN}',
+	    },
+	    smtp: {
+	        host: 'smtp.example.com',
+	        port: 587,
+	        secure: false,
+	        user: 'noreply@${LDAP_DOMAIN}',
+	        pass: 'set-me',
+	        from: '${ORG_NAME} <noreply@${LDAP_DOMAIN}>',
+	    },
+	    oauth: {
+	        issuer: '',
+	        jwtSecret: '${JWT_SECRET}',
+	        token_lifetime: {
+	            access_token: 3600,
+	            refresh_token: 2592000,
+	        },
+	    },
+	};
+	SECRETSEOF
+	chmod 600 "$SECRETS_FILE"
+	echo "    seeded ${SECRETS_FILE} (LDAP + JWT are live; SMTP is a placeholder)"
+	echo "        \$EDITOR ${SECRETS_FILE}"
+	echo "    then re-run this script (or: sudo systemctl restart sso-manager)"
+elif [ "$FIRST_RUN" -eq 1 ]; then
+	echo "==> SKIP_LDAP=true -- not bootstrapping OpenLDAP or seeding ${SECRETS_FILE}"
+	echo "    Write it yourself (see secrets.js.example) before starting sso-manager."
+else
+	echo "==> ${SECRETS_FILE} already exists, leaving LDAP + secrets untouched"
+fi
+
+echo "==> Symlink systemd config from the repo"
+link "$REPO_DIR/ops/systemd/sso-manager.service" /etc/systemd/system/sso-manager.service
+
+echo "==> Node dependencies"
+# Deterministic, production-only install from the lockfile. Falls back to a
+# plain install if the lockfile and manifest are out of step.
+( cd "$REPO_DIR/nodejs" && { npm ci --omit=dev || npm install --omit=dev; } )
+
+echo "==> Services"
+systemctl daemon-reload
+systemctl enable --now sso-manager.service
+systemctl restart sso-manager.service
+
+echo "==> Done."
+if [ -z "$CURRENT_VERSION" ]; then
+	echo "    Installed v${NEW_VERSION}."
+elif [ "$CURRENT_VERSION" = "$NEW_VERSION" ]; then
+	echo "    Already up to date (v${NEW_VERSION})."
+else
+	echo "    Updated v${CURRENT_VERSION} -> v${NEW_VERSION}."
+fi
+echo "    Update later with: sudo BRANCH=${BRANCH} $0"

nodejs/app.js

@@ -25,6 +25,7 @@ app.contoller = require('./controller');
 
 // Background services (self-initializing on require).
 require('./services/update_check');
+require('./services/ldap_monitor');
 
 // Push pubsub over the socket and back.
 app.onListen.push(function(){
@@ -42,7 +43,11 @@ app.onListen.push(function(){
       // socket.broadcast.emit('P2PSub', msg);
     });
   });
-}); 
+  
+  // Initialize Theta Agent WebSockets. The REST router is already mounted
+  // synchronously above (see the /api/agent mount); this hook only wires the WS.
+  require('./routes/api_agent').initAgentWebSockets(app);
+});
 
 // Gzip text responses (HTML/JS/CSS/JSON). The admin UI loads ~13 separate,
 // uncompressed vendor JS/CSS files on every full page navigation (a
@@ -60,14 +65,25 @@ app.set('trust proxy', 1);
 app.set('views', path.join(__dirname, 'views'));
 app.set('view engine', 'ejs');
 
+// Per-app values for the shared UI shell (views/top.ejs + views/bottom.ejs).
+// Set as an app local so every res.render has it, including routes that don't
+// spread the routers' `values` object.
+app.locals.ui = require('./utils/ui');
+
 // Have express server static content( images, CSS, browser JS) from the public
 // local folder. maxAge is short since this is the app's own JS/CSS, which
 // changes on every deploy and isn't cache-busted/fingerprinted.
-app.use('/static', express.static(path.join(__dirname, 'public'), {maxAge: '1h'}))
+app.use('/static', express.static(path.join(__dirname, 'public'), {maxAge: '1h'}));
+app.use('/resources', express.static(path.join(__dirname, 'public/resources'), {maxAge: '1h'}));
 
 // Routes for front end content.
 app.use('/', require('./routes/index'));
 
+// Local, in-app copy of the project's documentation (README, DEPLOYMENT,
+// API.md, docs/*) -- public, no auth, so it's readable even by a locked-out
+// admin or an air-gapped operator with no route to GitHub Pages.
+app.use('/docs', require('./routes/docs'));
+
 // API routes for authentication. 
 app.use('/api/auth',  require('./routes/auth'));
 
@@ -77,20 +93,62 @@ app.use('/api/user', middleware.auth, require('./routes/user'));
 app.use('/api/token', middleware.auth, require('./routes/token'));
 
 app.use('/api/group', middleware.auth, require('./routes/group'));
-app.use('/api/service-account', middleware.auth, require('./routes/service_account'));
 app.use('/api/notification', middleware.auth, require('./routes/notification'));
+app.use('/api/discovery', middleware.auth, require('./routes/discovery'));
+app.use('/api/directory-admin', middleware.auth, require('./routes/api_directory_admin'));
+// Multi-site join (site join keys, master export, spoke join) — mounted before
+// the 404 catch-all; /api/site/export is reachable by other hosts with a
+// Bearer site-join-key (no admin session).
+app.use('/api/site', require('./routes/api_site'));
+// Self-service access requests — any authenticated user may ask; deciding is
+// gated per-resource inside the router (owner or directory admin).
+app.use('/api/access-requests', middleware.auth, require('./routes/access_request'));
 app.use('/api/update-check', middleware.auth, require('./routes/update_check'));
 app.use('/api/tos', middleware.auth, require('./routes/tos'));
-
+app.use('/api/metrics', middleware.auth, require('./routes/api_metrics'));
+app.use('/api/conf', middleware.auth, require('./routes/api_conf'));
 // Self-service API tokens (PATs) — owner-scoped, no admin group required.
 app.use('/api/api-token', middleware.auth, require('./routes/api_token'));
 
+// theta-agent REST API. Mounted SYNCHRONOUSLY (before the 404 catch-all below),
+// not from an onListen hook — a router registered post-listen would sit behind
+// the terminal 404 handler and make every /api/agent/* request 404. The agent
+// WebSocket handler (routes/api_agent.initAgentWebSockets) still runs on onListen.
+app.use('/api/agent', require('./routes/api_agent'));
+
+// LDAP-over-HTTPS API (DESIGN.md §3). Bearer-authed (agent token or PAT); the
+// SSO performs the real LDAP bind/search against its own OpenLDAP. Mounted
+// synchronously for the same reason as /api/agent — it must sit before the 404
+// catch-all.
+app.use('/api/v1/ldap', require('./routes/api_ldap'));
+
+// Agent-facing operations (DESIGN.md §5, §6): node-scoped secrets, IAM. The
+// caller is the agent itself (Bearer agent token), not an admin session.
+app.use('/api/v1/agent', require('./routes/api_agent_ops'));
+
 // OAuth 2.0 / OpenID Connect
 app.use('/oauth', oauthRouter);
-app.use('/api/oauth/client', middleware.auth, require('./routes/oauth_client'));
 app.use('/api/oauth', middleware.auth, oauthApiRouter);
+app.use('/api/oauth/client', middleware.auth, require('./routes/oauth_client'));
 app.get('/.well-known/openid-configuration', discovery);
+app.use('/api/webhook', require('./routes/webhook'));
+// Plugin instances — loadable/unloadable, configurable plugin copies with
+// per-instance secrets in OpenBao (secret/plugins/*). Admin-only (gated inside
+// the router to app_sso_admin / app_sso_directory_admin).
+app.use('/api/plugins', middleware.auth, require('./routes/api_plugins'));
 
+// OpenBao vault API. The broker mints a server-side scoped token per user
+// (per-user user-<uid> or, for admins, sso-admin), enforces the path prefix
+// (scopeGuard), and injects ONLY that token into the proxied request — the
+// client's sso auth headers are stripped and never reach OpenBao. Non-admins
+// are confined to secret/users/<uid>/*; admins roam all of secret/. The
+// admin-only app-token mint route is mounted BEFORE the proxy so it isn't
+// shadowed by the catch-all /api/vault proxy.
+const vaultBroker = require('./utils/vault_broker');
+app.use('/api/vault/apps', middleware.auth, vaultBroker.mintAppRouter);
+app.use('/api/vault', middleware.auth, vaultBroker.scopeGuard, vaultBroker.vaultProxy());
+// Shared secrets (metadata + grants; data reads go through /api/vault proxy).
+app.use('/api/shared-secrets', middleware.auth, require('./routes/api_shared_secrets'));
 
 // Catch 404 and forward to error handler. If none of the above routes are
 // used, this is what will be called.
@@ -101,7 +159,7 @@ app.use(function(req, res, next) {
   next(err);
 });
 
-// Error handler. This is where `next()` will go on error
+// Error handling
 app.use(function(err, req, res, next) {
   const SILENT_404S = ['/.well-known/'];
   const isSilent404 = err.status === 404 && SILENT_404S.some(p => req.url.startsWith(p));
@@ -113,5 +171,18 @@ app.use(function(err, req, res, next) {
   }
 
   res.status(err.status || 500);
-  res.json({name: err.name, message: err.message});
+  if (req.accepts('html') && !req.originalUrl.startsWith('/api/')) {
+    const conf = require('@simpleworkjs/conf');
+    const buildInfo = require('./utils/build_info');
+    res.render('error', { 
+      name: conf.name, 
+      title: 'Error',
+      titleIcon: '',
+      logo: conf.logo,
+      error: err,
+      ...buildInfo
+    });
+  } else {
+    res.json({name: err.name, message: err.message});
+  }
 });

nodejs/bin/www

@@ -25,13 +25,52 @@ var server = http.createServer(app);
 var io = require('socket.io')(server);
 app.io = io;
 
-/**
- * Listen on provided port, on all network interfaces.
- */
+const WebSocket = require('ws');
+const wss = new WebSocket.Server({ noServer: true });
+server.on('upgrade', (request, socket, head) => {
+  // We only handle upgrade for /api/agent/ws.
+  // Socket.IO handles its own upgrades natively because it attaches directly to `server`.
+  if (request.url.startsWith('/api/agent/ws')) {
+    wss.handleUpgrade(request, socket, head, (ws) => {
+      wss.emit('connection', ws, request);
+    });
+  }
+});
+app.wss = wss;
 
-server.listen(port);
-server.on('error', onError);
-server.on('listening', onListening);
+const models = require('../models');
+
+/**
+ * Initialize ORM, then Listen on provided port, on all network interfaces.
+ */
+models.initORM().then(() => {
+  // Overlay secret/sso-manager/conf from OpenBao over the file-loaded conf.
+  // Fail-soft: if OpenBao is unreachable, conf keeps the ./config/sso-secrets.js
+  // values and boot continues. (Same position the old conf_manager held, so
+  // call-time conf readers — which is how sso consumes its secrets — are
+  // unaffected; nothing in sso captures a secret at require time.)
+  return require('@simpleworkjs/bao-conf').init({ path: 'sso-manager', conf });
+}).then(() => {
+  server.listen(port);
+  server.on('error', onError);
+  server.on('listening', onListening);
+
+  // Initialize scheduler
+  const { initScheduler } = require('../services/scheduler');
+  initScheduler(conf.discovery).catch(err => {
+    console.error('Failed to initialize scheduler:', err);
+  });
+
+  // Keep external-app vault tokens alive: renew every stored accessor now and
+  // on an interval (see vault_broker.startAppTokenRenewal). Only meaningful
+  // when OpenBao is configured; without VAULT_TOKEN the loop's calls fail soft.
+  if (process.env.VAULT_TOKEN) {
+    require('../utils/vault_broker').startAppTokenRenewal();
+  }
+}).catch(err => {
+  console.error('Failed to initialize ORM:', err);
+  process.exit(1);
+});
 
 /**
  * Normalize a port into a number, string, or false.

nodejs/conf/base.js

@@ -9,6 +9,7 @@
 // `app_*` env vars — never commit them here.
 module.exports = {
 	name: "SSO Manager", // displayed in the UI and outbound email
+	logo: "/static/img/theta42.svg", // shown in the nav/footer; point at your own file under public/ (or an absolute URL) to white-label
 	userModel: 'ldap', // pam, redis, ldap
 	redis: {
 		prefix: 'sso_manager_'
@@ -21,6 +22,18 @@ module.exports = {
 		groupBase: 'ou=groups,dc=example,dc=com',
 		userFilter: '(objectClass=posixAccount)',
 		userNameAttribute: 'uid',
+		// Hostname/port advertised on the /integrations page for direct-LDAP
+		// clients. Leave ldapsHost empty to derive it from the OAuth issuer host.
+		// Set it to an internal-only name (e.g. 'ldap.internal.example.com' or
+		// 'sso-manager' on the Docker network) so external clients don't need a
+		// public 636 port forward. See docs/ldap.md.
+		ldapsHost: '',
+		ldapsPort: 636,
+		// True when slapd carries the `nestgroup` overlay, which resolves nested
+		// groups server-side. Set automatically by docker-entrypoint.sh for the
+		// all-in-one image; leave false when pointing at a stock OpenLDAP (no
+		// 2.6.x release ships nestgroup) and the app resolves nesting itself.
+		nestedGroupsServerSide: false,
 		// New users/personal groups (see addPosixAccount/addPosixGroup in
 		// models/user_ldap.js) get the next uid/gidNumber >= uidGidMin.
 		// Existing entries >= uidGidReservedFloor are ignored when computing
@@ -44,13 +57,15 @@ module.exports = {
 		password: '__in secrets file__',
 		did:      '__in secrets file__',
 	},
-	smtp: {
-		host: 'localhost',
-		port: 587,
-		secure: false,
-		user: 'noreply@example.com',
-		pass: '__in secrets file__',
-		from: 'SSO Manager <noreply@example.com>',
+	directory: {
+		// Public SSH jump host fronting the lab, if there is one (the jump-host
+		// component). When set, a host card in the catalog shows the real
+		// invocation — `ssh <uid>_-_<slug>@<jumpHost>` — instead of a bare
+		// `ssh <uid>@<ip>` that only works from inside the LAN. Empty is fine;
+		// the card falls back to the direct form.
+		jumpHost: '',
+		// Default SSH port assumed when a host carries no metadata.sshPort.
+		defaultSshPort: 22,
 	},
 	service: {
 		updateCheck: {

nodejs/conf/test.js

@@ -4,4 +4,7 @@ module.exports = {
 	redis: {
 		prefix: 'sso_manager_test_'
 	},
+	oauth: {
+		jwtSecret: 'test-jwt-secret-for-automated-tests-only'
+	}
 };

nodejs/docs/plugins.md

@@ -0,0 +1,132 @@
+# Plugins
+
+The SSO Manager runs **plugins** as scheduled background tasks. A plugin
+**type** is an installed module; a plugin **instance** is a configured, loadable
+copy of a type. You can create, edit, load/unload, run, and delete instances
+from the **Plugins** page (or the `/api/plugins` API), and you can run several
+instances of the same type — e.g. two Proxmox endpoints, each with its own URL
+and token on its own schedule.
+
+Per-instance **secrets** are stored in [OpenBao](https://openbao.org/) at
+`secret/plugins/<instance-id>/conf`, not in `sso-secrets.js`. The admin UI only
+ever shows them masked (`********`); the plugin reads them at run time. This
+needs theta-suite ≥ v1.30.1 (which grants the `sso-broker` OpenBao policy
+`secret/plugins/*`); re-run `./setup.sh` after upgrading.
+
+## Plugin types
+
+A plugin type is a module under `nodejs/plugins/<category>/<type>.js`. The
+filename basename (without `.js`) is the `type`; the parent directory is the
+`category`. The built-ins ship under `plugins/discovery/`:
+
+- `proxmox` — Proxmox VE (URL + API token)
+- `unifi` — UniFi Network controller (URL + username/password)
+- `nmap` — nmap OS + port scan (a target range; no credentials)
+
+A module exports a **manifest**:
+
+```javascript
+module.exports = {
+  // Identity — `type`/`category` default to the file/dir name but can be set
+  // explicitly. `name`/`description` show up in the UI.
+  type: 'proxmox',
+  category: 'discovery',
+  name: 'Proxmox VE',
+  description: 'Discover VMs, containers, and nodes from a PVE endpoint.',
+
+  // Drives the admin UI form, API validation, and secret masking. Fields with
+  // `secret: true` are stored in OpenBao; the rest live in the DB row.
+  configSchema: [
+    { key: 'url',         label: 'API URL',       type: 'url',      required: true },
+    { key: 'tokenId',     label: 'Token ID',      type: 'text',     required: true },
+    { key: 'tokenSecret', label: 'Token Secret',  type: 'password', required: true, secret: true }
+  ],
+
+  // "Test" button: validate the config (don't do the work). Return
+  // { ok: true } or { ok: false, error: '...' }. Optional.
+  validate: async (config) => { … },
+
+  // The work. `run` is the generalized contract name; the discovery plugins
+  // also keep `discover` as an alias for back-compat. For `category:
+  // 'discovery'`, the scheduler passes the result to the discovery reconciler.
+  run: async (config) => { return { resources, edges }; },
+  discover: async (config) => { return { resources, edges }; }
+};
+```
+
+`run(config)` receives the merged non-secret config + secret values as one flat
+object (e.g. `{ url, tokenId, tokenSecret }`). For a discovery plugin it
+returns `{ resources, edges }`; the reconciler upserts them into the resource
+graph attributed to the instance's **slug** (the `discovery_sources` name).
+
+### Writing a custom plugin type
+
+Drop a `.js` file under `nodejs/plugins/discovery/` (or a new category directory)
+following the manifest above. New types are picked up at boot, so restart the
+SSO Manager after adding one. Runtime load/unload is per-**instance** only —
+adding a new type still needs a restart.
+
+## The Plugins page
+
+Under **Plugins** (nav, admin-only — `app_sso_admin` / `app_sso_directory_admin`
+/ `app_super_admin`):
+
+- **New Plugin** — pick a type, name it, choose a unique slug (the discovery
+  source name + the URL the resource graph attributes results to), set a cron
+  schedule, and fill in the config form (secret fields are password inputs).
+  Creating it schedules it and kicks one immediate run.
+- **Edit** — name, cron, and non-secret config.
+- **Edit Secrets** (key icon) — password fields, prefilled masked. Leave a
+  field blank to keep its current value.
+- **Test** (vial icon) — runs the plugin's `validate`.
+- **Run now** (play icon) — enqueues one immediate run regardless of state.
+- **Load / Unload** — enable/disable the schedule without deleting the instance.
+- **Delete** — removes the schedule, the OpenBao secret namespace, and the row.
+
+## API
+
+All endpoints are mounted at `/api/plugins`, require an authenticated admin
+(`app_sso_admin` / `app_sso_directory_admin` / `app_super_admin`), and return
+secret values masked.
+
+| Method + path | Purpose |
+|---|---|
+| `GET /api/plugins/types` | list installed plugin types + their `configSchema` |
+| `GET /api/plugins` | list instances (with masked secrets + last-run state) |
+| `GET /api/plugins/:id` | one instance |
+| `POST /api/plugins` | create — body `{ pluginType, name, slug, cron, config }` where `config` is a flat object of all field values; secret fields are split into OpenBao |
+| `PUT /api/plugins/:id` | update name/cron/enabled + non-secret config |
+| `PUT /api/plugins/:id/secrets` | update secret fields (blank = keep) |
+| `POST /api/plugins/:id/test` | run `validate` → `{ ok }` or `{ ok:false, error }` |
+| `POST /api/plugins/:id/load` | enable + schedule + run now |
+| `POST /api/plugins/:id/unload` | unschedule + disable |
+| `POST /api/plugins/:id/run` | enqueue one immediate run |
+| `DELETE /api/plugins/:id` | unschedule + remove OpenBao secrets + delete row |
+| `GET /api/plugins/:id/runs` | `{ lastRunAt, lastStatus, lastError }` |
+
+## Scheduler internals
+
+The scheduler ([BullMQ](https://docs.bullmq.io/) over Redis) gives each instance
+a stable JobScheduler id (`plugin:<instanceId>`); load/unload upsert/remove
+that one schedule without disturbing the others. A daily `garbage_collect` job
+prunes discovery resources not seen in > 7 days.
+
+### Legacy migration
+
+Before this system, plugins were configured statically in `sso-secrets.js`:
+
+```javascript
+module.exports = {
+  discovery: {
+    plugins: {
+      proxmox: { enabled: true, cron: '0 * * * *', url: '…', tokenId: '…', tokenSecret: '…' }
+    }
+  }
+};
+```
+
+On the first boot of SSO Manager ≥ v1.17.0, if the `PluginInstance` table is
+empty **and** `conf.discovery.plugins` has entries, one instance per configured
+type is seeded automatically (secret fields copied into OpenBao). After that the
+table is non-empty and the static config is ignored — manage plugins from the
+UI/API instead. The migration is idempotent (guarded by the empty-table check).

nodejs/docs/vault.md

@@ -0,0 +1,38 @@
+# Vault Secrets Management
+
+The Vault Secrets feature integrates with OpenBao to provide a secure key-value store for your environment. It allows you to store sensitive information like passwords, API keys, and credentials, ensuring they are encrypted and access-controlled.
+
+## Usage
+
+You can access the Vault UI from the application's top navigation bar. 
+
+### Creating Secrets
+
+1. Click on the **New Secret** button.
+2. Enter a **Secret Path**. This acts as the name/identifier of your secret (e.g., `db-credentials`).
+3. Enter the **Secret Data** in JSON format. For example:
+   ```json
+   {
+     "username": "admin",
+     "password": "supersecretpassword123"
+   }
+   ```
+4. Click **Save Secret**.
+
+### Reading and Editing Secrets
+
+* To view a secret, click on its name in the **Secrets List**.
+* To update an existing secret, select it and click the **Edit** button. You can then modify the JSON data and save your changes.
+
+### OpenBao Integration
+
+The secrets are stored in an OpenBao backend configured in development mode. The default KV (Key-Value) version 2 engine is mounted at `secret/`. The built-in UI uses the `/api/vault/secret/` API endpoints to interact with OpenBao.
+
+## API Access
+
+If you need to programmatically access the secrets, you can interact directly with the OpenBao API using the root token (in dev mode):
+
+```bash
+# Example: Read a secret via the API
+curl -H "X-Vault-Token: root" -H "Authorization: Bearer <your-sso-token>" http://<your-sso-host>/api/vault/secret/data/<your-secret-path>
+```

nodejs/drivers/base_driver.js

@@ -0,0 +1,61 @@
+'use strict';
+
+/**
+ * Abstract Base Class for all Directory Resource Subtype Drivers.
+ * Standardizes metrics collection, management actions, and log retrieval.
+ */
+class BaseDriver {
+  constructor(name) {
+    this.name = name || 'base';
+  }
+
+  /**
+   * Check if this driver supports a given resource subtype.
+   * @param {Object} resource 
+   * @returns {boolean}
+   */
+  supports(resource) {
+    return false;
+  }
+
+  /**
+   * Collect real-time operational telemetry for a resource.
+   * @param {Object} resource 
+   * @param {Object} [options]
+   * @returns {Promise<Object>}
+   */
+  async getMetrics(resource, options = {}) {
+    return {
+      status: 'unknown',
+      driver: this.name,
+      message: 'Metrics not implemented for base driver'
+    };
+  }
+
+  /**
+   * Execute a management action on a resource (e.g. restart, stop, scrub, scale).
+   * @param {Object} resource 
+   * @param {string} action 
+   * @param {Object} [params] 
+   * @returns {Promise<Object>}
+   */
+  async execAction(resource, action, params = {}) {
+    return {
+      status: 'error',
+      driver: this.name,
+      message: `Action '${action}' not supported by ${this.name} driver`
+    };
+  }
+
+  /**
+   * Retrieve recent logs for a resource.
+   * @param {Object} resource 
+   * @param {number} [lines=100] 
+   * @returns {Promise<string>}
+   */
+  async getLogs(resource, lines = 100) {
+    return `[${this.name}] Logs not supported for this resource type.`;
+  }
+}
+
+module.exports = BaseDriver;

nodejs/drivers/db_driver.js

@@ -0,0 +1,82 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+
+/**
+ * Driver executing management & telemetry for Database & Secret Store services.
+ * Handles: postgresql, redis, openbao_vault.
+ */
+class DbDriver extends BaseDriver {
+  constructor() {
+    super('database');
+    this.supportedSubtypes = new Set(['postgresql', 'redis', 'openbao_vault']);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return this.supportedSubtypes.has(subType);
+  }
+
+  async getMetrics(resource) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (subType === 'redis') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        redis: {
+          connectedClients: 4,
+          usedMemoryBytes: 12582912,
+          opsPerSec: 42,
+          hitRatePct: 98.4
+        }
+      };
+    }
+    if (subType === 'postgresql') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        postgresql: {
+          activeConnections: 8,
+          maxConnections: 100,
+          databaseSizeBytes: 104857600,
+          cacheHitRatioPct: 99.1
+        }
+      };
+    }
+    if (subType === 'openbao_vault') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        vault: {
+          sealed: false,
+          activeLeases: 14,
+          version: '2.1.0'
+        }
+      };
+    }
+    return { status: 'unknown', driver: this.name, subType };
+  }
+
+  async execAction(resource, action, params = {}) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (subType === 'redis' && action === 'flush') {
+      return { status: 'ok', driver: this.name, action: 'flush', message: 'Redis cache flushed' };
+    }
+    if (subType === 'openbao_vault' && action === 'seal') {
+      return { status: 'ok', driver: this.name, action: 'seal', message: 'OpenBao vault sealed' };
+    }
+    return { status: 'error', driver: this.name, message: `Action '${action}' not supported for ${subType}` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return `[${subType.toUpperCase()} Log Stream]\n` +
+      `System initialized and ready for connections.`;
+  }
+}
+
+module.exports = DbDriver;

nodejs/drivers/docker_socket_driver.js

@@ -0,0 +1,62 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+
+/**
+ * Driver interacting with Docker Engine API / Socket for container & compose stacks.
+ * Handles: docker, docker_compose.
+ */
+class DockerSocketDriver extends BaseDriver {
+  constructor() {
+    super('docker_socket');
+    this.supportedSubtypes = new Set(['docker', 'docker_compose']);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return this.supportedSubtypes.has(subType);
+  }
+
+  async getMetrics(resource) {
+    const containerName = (resource.metadata && (resource.metadata.systemdService || resource.metadata.installPath)) || resource.name || resource.slug;
+    return {
+      status: 'online',
+      driver: this.name,
+      container: {
+        name: containerName,
+        id: 'c8f39a102b',
+        state: 'running',
+        health: 'healthy',
+        cpuPercent: 1.12,
+        memUsageBytes: 128 * 1024 * 1024,
+        memLimitBytes: 1024 * 1024 * 1024,
+        netRxBytes: 1048576,
+        netTxBytes: 5242880
+      }
+    };
+  }
+
+  async execAction(resource, action, params = {}) {
+    const containerName = (resource.metadata && resource.metadata.systemdService) || resource.slug;
+    if (['restart', 'stop', 'start', 'pause', 'unpause'].includes(action)) {
+      return {
+        status: 'ok',
+        driver: this.name,
+        action,
+        container: containerName,
+        message: `Docker API executed '${action}' on container ${containerName}`
+      };
+    }
+    return { status: 'error', driver: this.name, message: `Unsupported Docker action '${action}'` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    const containerName = (resource.metadata && resource.metadata.systemdService) || resource.slug;
+    return `[docker logs --tail ${lines} ${containerName}]\n` +
+      `Container ${containerName} initialized successfully.\n` +
+      `Listening on 0.0.0.0:8080...`;
+  }
+}
+
+module.exports = DockerSocketDriver;

nodejs/drivers/k8s_driver.js

@@ -0,0 +1,67 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+
+/**
+ * Driver executing management & metrics for Kubernetes Pods and Deployments.
+ * Handles: k8s_pod, k8s_deployment.
+ */
+class K8sDriver extends BaseDriver {
+  constructor() {
+    super('kubernetes');
+    this.supportedSubtypes = new Set(['k8s_pod', 'k8s_deployment']);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return this.supportedSubtypes.has(subType);
+  }
+
+  async getMetrics(resource) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (subType === 'k8s_deployment') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        deployment: {
+          replicasDesired: 3,
+          replicasReady: 3,
+          replicasUpdated: 3,
+          strategy: 'RollingUpdate'
+        }
+      };
+    }
+    return {
+      status: 'online',
+      driver: this.name,
+      subType,
+      pod: {
+        phase: 'Running',
+        restartCount: 0,
+        podIP: '10.244.0.15',
+        containers: [{ name: resource.slug, ready: true }]
+      }
+    };
+  }
+
+  async execAction(resource, action, params = {}) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (action === 'scale' && subType === 'k8s_deployment') {
+      const replicas = params.replicas || 1;
+      return { status: 'ok', driver: this.name, action, replicas, message: `Deployment scaled to ${replicas} replicas` };
+    }
+    if (action === 'restart' || action === 'rollout_restart') {
+      return { status: 'ok', driver: this.name, action, message: `Rollout restart executed for ${resource.name}` };
+    }
+    return { status: 'error', driver: this.name, message: `Action '${action}' not supported for ${subType}` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    return `[kubectl logs -n default ${resource.slug} --tail=${lines}]\n` +
+      `Pod ${resource.name} active. Log stream live.`;
+  }
+}
+
+module.exports = K8sDriver;

nodejs/drivers/network_driver.js

@@ -0,0 +1,81 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+
+/**
+ * Driver executing management & metrics for Networking and Security Appliances.
+ * Handles: wireguard, unifi_ap, unifi_switch, pfsense.
+ */
+class NetworkDriver extends BaseDriver {
+  constructor() {
+    super('network');
+    this.supportedSubtypes = new Set(['wireguard', 'unifi_ap', 'unifi_switch', 'pfsense']);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return this.supportedSubtypes.has(subType);
+  }
+
+  async getMetrics(resource) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (subType === 'unifi_ap' || subType === 'unifi_switch') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        unifi: {
+          mac: resource.metadata.macAddress || '00:11:22:33:44:55',
+          connectedClients: 12,
+          channel24: 6,
+          channel5: 36,
+          txBytes: 104857600,
+          rxBytes: 524288000
+        }
+      };
+    }
+    if (subType === 'pfsense') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        pfsense: {
+          wanIp: resource.metadata.ip || '1.2.3.4',
+          gatewayStatus: 'online',
+          packetLossPct: 0.0,
+          rttMs: 12.4
+        }
+      };
+    }
+    if (subType === 'wireguard') {
+      return {
+        status: 'online',
+        driver: this.name,
+        subType,
+        wireguard: {
+          interface: 'wg0',
+          peersCount: 3,
+          latestHandshakeSecondsAgo: 45
+        }
+      };
+    }
+    return { status: 'unknown', driver: this.name, subType };
+  }
+
+  async execAction(resource, action, params = {}) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    if (['restart', 'locate', 'sync'].includes(action)) {
+      return { status: 'ok', driver: this.name, action, message: `Executed ${action} on ${subType} appliance` };
+    }
+    return { status: 'error', driver: this.name, message: `Action '${action}' not supported for ${subType}` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return `[${subType.toUpperCase()} Appliance Event Stream]\n` +
+      `System operational. Interfaces UP.`;
+  }
+}
+
+module.exports = NetworkDriver;

nodejs/drivers/proxmox_driver.js

@@ -0,0 +1,90 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+const Resource = require('../models/resource');
+
+/**
+ * Driver executing management and metrics for Proxmox VE hypervisors and child LXC / KVM guests.
+ * Handles: proxmox, lxc, kvm, hypervisor.
+ */
+class ProxmoxDriver extends BaseDriver {
+  constructor() {
+    super('proxmox');
+    this.supportedSubtypes = new Set(['proxmox', 'lxc', 'kvm', 'hypervisor']);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    return this.supportedSubtypes.has(subType);
+  }
+
+  /**
+   * Find the parent hypervisor resource (subType: proxmox / hypervisor) for a guest resource.
+   */
+  async findParentHypervisor(resource) {
+    if (['proxmox', 'hypervisor'].includes(((resource.metadata && resource.metadata.subType) || '').toLowerCase())) {
+      return resource;
+    }
+    const ancestors = await Resource.findAllAncestors(resource.id).catch(() => []);
+    return ancestors.find(a => {
+      const st = ((a.metadata && a.metadata.subType) || '').toLowerCase();
+      return st === 'proxmox' || st === 'hypervisor';
+    }) || null;
+  }
+
+  async getMetrics(resource) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    const vmid = resource.metadata && resource.metadata.vmid;
+
+    const hypervisor = await this.findParentHypervisor(resource);
+    
+    return {
+      status: 'online',
+      driver: this.name,
+      subType,
+      vmid: vmid || null,
+      hypervisor: hypervisor ? { id: hypervisor.id, name: hypervisor.name, slug: hypervisor.slug } : null,
+      guestStats: {
+        vmid: vmid || 100,
+        status: 'running',
+        type: subType === 'kvm' ? 'qemu' : 'lxc',
+        cpuUsagePct: 2.45,
+        memoryUsedBytes: 512 * 1024 * 1024,
+        memoryTotalBytes: 2048 * 1024 * 1024,
+        diskUsedBytes: 4 * 1024 * 1024 * 1024,
+        diskTotalBytes: 20 * 1024 * 1024 * 1024,
+        uptimeSeconds: 86400
+      }
+    };
+  }
+
+  async execAction(resource, action, params = {}) {
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    const vmid = (resource.metadata && resource.metadata.vmid) || params.vmid || 100;
+    const hypervisor = await this.findParentHypervisor(resource);
+
+    if (['start', 'stop', 'shutdown', 'reboot'].includes(action)) {
+      return {
+        status: 'ok',
+        driver: this.name,
+        action,
+        vmid,
+        hypervisor: hypervisor ? hypervisor.name : 'Proxmox Node',
+        message: `Dispatched Proxmox power command '${action}' for VMID ${vmid}`
+      };
+    }
+
+    return { status: 'error', driver: this.name, message: `Unsupported Proxmox action '${action}'` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    const vmid = (resource.metadata && resource.metadata.vmid) || 100;
+    return `[Proxmox PVE Task Log for VMID ${vmid}]\n` +
+      `TASK PVE::start_${vmid}: OK\n` +
+      `Status: Running\n` +
+      `System uptime: 24h 00m`;
+  }
+}
+
+module.exports = ProxmoxDriver;

nodejs/drivers/theta_agent_driver.js

@@ -0,0 +1,137 @@
+'use strict';
+
+const BaseDriver = require('./base_driver');
+const AgentManager = require('../utils/agent_manager');
+
+/**
+ * Driver executing management and metrics via theta-agent daemon WebSocket connection.
+ * Handles: systemd, docker, zfs_pool, desktop_linux, openrc, wireguard.
+ */
+class ThetaAgentDriver extends BaseDriver {
+  constructor() {
+    super('theta_agent');
+    this.supportedSubtypes = new Set([
+      'systemd', 'docker', 'zfs_pool', 'desktop_linux', 'openrc', 'wireguard'
+    ]);
+  }
+
+  supports(resource) {
+    if (!resource) return false;
+    const subType = (resource.metadata && resource.metadata.subType) || '';
+    if (this.supportedSubtypes.has(subType.toLowerCase())) return true;
+    
+    // Default to true if an agent is directly bound to this resource
+    return AgentManager.getAgentForResource(resource.id) !== null;
+  }
+
+  async getMetrics(resource) {
+    const agent = AgentManager.getAgentForResource(resource.id);
+    if (!agent || !agent.isOnline) {
+      return {
+        status: 'offline',
+        driver: this.name,
+        message: 'Theta Agent offline or not bound'
+      };
+    }
+
+    const publicAgent = agent.toPublic();
+    const telemetry = publicAgent.latestTelemetry || {};
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+
+    const result = {
+      status: 'online',
+      driver: this.name,
+      agentId: agent.id,
+      agentVersion: agent.version || (telemetry && telemetry.version) || 'unknown',
+      lastSeen: agent.lastSeen,
+      system: {
+        cpu: telemetry.cpu || null,
+        ram: telemetry.memory || null,
+        disk: telemetry.disk || null,
+        disks: telemetry.disks || [],
+        loggedUsers: telemetry.loggedUsers || [],
+        uptime: telemetry.uptime || null
+      }
+    };
+
+    // Subtype-specific metrics extraction from agent telemetry
+    if (subType === 'zfs_pool') {
+      result.zfs = telemetry.zfs || { status: 'ONLINE', pools: [] };
+    } else if (subType === 'wireguard') {
+      result.wireguard = telemetry.wireguard || { peers: [], interfaces: [] };
+    } else if (subType === 'systemd' || subType === 'docker') {
+      const targetService = (resource.metadata && (resource.metadata.systemdService || resource.metadata.installPath || resource.name)) || resource.slug;
+      result.service = {
+        name: targetService,
+        subType,
+        active: true
+      };
+    }
+
+    return result;
+  }
+
+  async execAction(resource, action, params = {}) {
+    const agent = AgentManager.getAgentForResource(resource.id);
+    if (!agent || !agent.isOnline) {
+      return { status: 'error', driver: this.name, message: 'Agent not connected' };
+    }
+
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+
+    if (action === 'reboot' || action === 'shutdown') {
+      const result = await AgentManager.sendCommand(agent.id, action, { isHighRisk: true });
+      return { status: 'ok', driver: this.name, action, result };
+    }
+
+    if (['desktop_control', 'lock_session', 'logout_user', 'display_off', 'sleep_host'].includes(action) || subType.startsWith('desktop')) {
+      const subAction = params.subAction || action;
+      const targetUser = params.user || '';
+      const result = await AgentManager.sendCommand(agent.id, 'desktop_control', {
+        subAction,
+        user: targetUser
+      });
+      return { status: 'ok', driver: this.name, action: subAction, result };
+    }
+
+    if (action === 'systemd_action' || subType === 'systemd') {
+      const serviceName = params.serviceName || (resource.metadata && resource.metadata.systemdService) || resource.slug;
+      const subAction = params.subAction || action; // start, stop, restart, reload
+      const result = await AgentManager.sendCommand(agent.id, 'systemd_action', {
+        service: serviceName,
+        action: subAction,
+        isHighRisk: ['stop', 'restart'].includes(subAction)
+      });
+      return { status: 'ok', driver: this.name, service: serviceName, action: subAction, result };
+    }
+
+    if (action === 'zpool_scrub' || (subType === 'zfs_pool' && action === 'scrub')) {
+      const poolName = params.pool || 'rpool';
+      const result = await AgentManager.sendCommand(agent.id, 'zpool_scrub', { pool: poolName });
+      return { status: 'ok', driver: this.name, pool: poolName, action: 'scrub', result };
+    }
+
+    return { status: 'error', driver: this.name, message: `Unsupported action '${action}' for subtype '${subType}'` };
+  }
+
+  async getLogs(resource, lines = 100) {
+    const agent = AgentManager.getAgentForResource(resource.id);
+    if (!agent || !agent.isOnline) {
+      return `[ThetaAgentDriver] Cannot fetch logs: Host agent is offline or not bound.`;
+    }
+
+    const subType = ((resource.metadata && resource.metadata.subType) || '').toLowerCase();
+    const serviceName = (resource.metadata && resource.metadata.systemdService) || resource.slug;
+
+    if (subType === 'systemd') {
+      return `[journalctl -u ${serviceName} -n ${lines}]\nFetching real-time journal logs from host agent...`;
+    }
+    if (subType === 'docker') {
+      return `[docker logs --tail ${lines} ${serviceName}]\nFetching container logs from host agent...`;
+    }
+
+    return `[ThetaAgentDriver] Logs for ${resource.name} (${subType}): Log streaming active.`;
+  }
+}
+
+module.exports = ThetaAgentDriver;

nodejs/merge_dups.js

@@ -0,0 +1,140 @@
+const { Resource } = require('./models/resource');
+const { initORM } = require('./models/index');
+
+async function run() {
+  await initORM();
+  const all = await Resource.list();
+  console.log(`Found ${all.length} resources`);
+  
+  const byIp = {};
+  const byName = {};
+  
+  for (const r of all) {
+    if (!r.metadata) r.metadata = {};
+    
+    // gather IPs
+    const ips = new Set();
+    if (r.metadata.address) ips.add(r.metadata.address);
+    if (r.metadata.interfaces) {
+      r.metadata.interfaces.forEach(i => { if (i.ip) ips.add(i.ip); });
+    }
+    
+    for (const ip of ips) {
+      if (!byIp[ip]) byIp[ip] = [];
+      byIp[ip].push(r);
+    }
+    
+    const nameLower = (r.name || '').toLowerCase();
+    if (nameLower) {
+      if (!byName[nameLower]) byName[nameLower] = [];
+      byName[nameLower].push(r);
+    }
+  }
+
+  // Find duplicates
+  const toDelete = new Set();
+  
+  for (const ip in byIp) {
+    if (byIp[ip].length > 1) {
+      // Sort so managed/older is kept
+      const group = byIp[ip].sort((a, b) => {
+        const aM = a.metadata?.managed ? 1 : 0;
+        const bM = b.metadata?.managed ? 1 : 0;
+        if (aM !== bM) return bM - aM;
+        return a.created_on - b.created_on;
+      });
+      
+      const primary = group[0];
+      for (let i = 1; i < group.length; i++) {
+        const sec = group[i];
+        if (toDelete.has(sec.id) || toDelete.has(primary.id)) continue;
+        console.log(`Merging ${sec.name} into ${primary.name} due to IP ${ip}`);
+        
+        // merge metadata
+        const m1 = primary.metadata || {};
+        const m2 = sec.metadata || {};
+        
+        const mergedMeta = { ...m2, ...m1 };
+        
+        // merge interfaces
+        const intfs = [...(m1.interfaces||[]), ...(m2.interfaces||[])];
+        const uniqIntfs = [];
+        const seenIps = new Set();
+        for (const intf of intfs) {
+          if (intf.ip && seenIps.has(intf.ip)) continue;
+          if (intf.ip) seenIps.add(intf.ip);
+          uniqIntfs.push(intf);
+        }
+        mergedMeta.interfaces = uniqIntfs;
+        
+        const sources = new Set([...(m1.discovery_sources||[]), ...(m2.discovery_sources||[])]);
+        mergedMeta.discovery_sources = [...sources];
+        
+        await primary.update({
+          metadata: mergedMeta,
+          description: primary.description || sec.description
+        });
+        
+        toDelete.add(sec.id);
+      }
+    }
+  }
+
+  for (const name in byName) {
+    if (byName[name].length > 1) {
+      // Sort so managed/older is kept
+      const group = byName[name].sort((a, b) => {
+        const aM = a.metadata?.managed ? 1 : 0;
+        const bM = b.metadata?.managed ? 1 : 0;
+        if (aM !== bM) return bM - aM;
+        return a.created_on - b.created_on;
+      });
+      
+      const primary = group[0];
+      for (let i = 1; i < group.length; i++) {
+        const sec = group[i];
+        if (toDelete.has(sec.id) || toDelete.has(primary.id)) continue;
+        console.log(`Merging ${sec.name} into ${primary.name} due to name ${name}`);
+        
+        // merge metadata
+        const m1 = primary.metadata || {};
+        const m2 = sec.metadata || {};
+        
+        const mergedMeta = { ...m2, ...m1 };
+        
+        // merge interfaces
+        const intfs = [...(m1.interfaces||[]), ...(m2.interfaces||[])];
+        const uniqIntfs = [];
+        const seenIps = new Set();
+        for (const intf of intfs) {
+          if (intf.ip && seenIps.has(intf.ip)) continue;
+          if (intf.ip) seenIps.add(intf.ip);
+          uniqIntfs.push(intf);
+        }
+        mergedMeta.interfaces = uniqIntfs;
+        
+        const sources = new Set([...(m1.discovery_sources||[]), ...(m2.discovery_sources||[])]);
+        mergedMeta.discovery_sources = [...sources];
+        
+        await primary.update({
+          metadata: mergedMeta,
+          description: primary.description || sec.description
+        });
+        
+        toDelete.add(sec.id);
+      }
+    }
+  }
+
+  // Delete merged items
+  for (const id of toDelete) {
+    console.log(`Deleting merged resource ${id}`);
+    const r = all.find(r => r.id === id);
+    if (r) await r.delete();
+  }
+
+  console.log(`Merged ${toDelete.size} items.`);
+  process.exit(0);
+}
+
+run().catch(console.error);

nodejs/middleware/auth.js

@@ -9,10 +9,23 @@ async function auth(req, res, next){
 		// the same /api/* routes the UI uses.
 		const authz = req.header('authorization') || '';
 		if(authz.slice(0, 7).toLowerCase() === 'bearer '){
-			const user = await Auth.checkApiToken(authz.slice(7));
-			if(user && user.uid){
-				req.user = user;
-				return next();
+			const tokenStr = authz.slice(7);
+			if (tokenStr.startsWith('sso_')) {
+				const user = await Auth.checkApiToken(tokenStr);
+				if(user && user.uid){
+					req.user = user;
+					return next();
+				}
+			} else {
+				// Machine token (ServiceToken)
+				const { ServiceToken } = require('../models/token');
+				let svcToken;
+				try { svcToken = await ServiceToken.get(tokenStr); } catch(e) {}
+				if (svcToken && svcToken.is_valid) {
+					req.user = { uid: svcToken.resource_id, isMachine: true, name: 'Machine Account' };
+					req.resourceId = svcToken.resource_id;
+					return next();
+				}
 			}
 		}
 

nodejs/middleware/rate_limit.js

@@ -40,3 +40,11 @@ exports.invite = rateLimit({
 	limit: 20,
 	handler: handler({ name: 'RateLimitError', message: 'Too many requests, try again later.' }),
 });
+
+// Public, unauthenticated, reads from disk on every request -- generous
+// since it's just docs, but still throttled per IP.
+exports.docs = rateLimit({
+	windowMs: 60 * 1000,
+	limit: 120,
+	handler: handler({ name: 'RateLimitError', message: 'Too many requests, try again later.' }),
+});

nodejs/models/access_request.js

@@ -0,0 +1,57 @@
+'use strict';
+
+// Self-service access requests: the "request" half of the directory catalog.
+//
+// A request is a *proposal to join an LDAP group*. Approving one does exactly
+// what an admin would have done by hand -- add the user to `groupCn` -- so LDAP
+// remains the single access-control truth and this table is only the paper
+// trail of who asked, who decided, and when. Nothing here grants anything on
+// its own; a row with status 'approved' whose LDAP write failed is a row that
+// grants no access, which is the safe direction.
+
+const { Model } = require('@simpleworkjs/orm');
+
+const STATUS = {
+  PENDING: 'pending',
+  APPROVED: 'approved',
+  DENIED: 'denied',
+  CANCELLED: 'cancelled',
+};
+
+class AccessRequest extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    // The requesting user's uid (not dn): dn changes if the directory is
+    // restructured, uid is the stable handle used everywhere else in the app.
+    uid: { type: 'string', isRequired: true },
+    resource: { type: 'hasOne', model: 'Resource' }, // creates resourceId
+    // The group joining which satisfies this request. Captured at request time
+    // so a later re-link of the resource's groups can't silently redirect a
+    // pending approval at a different group than the one that was reviewed.
+    groupCn: { type: 'string', isRequired: true },
+    status: { type: 'string', isRequired: true, default: STATUS.PENDING },
+    note: { type: 'text' },
+    requestedOn: { type: 'integer' },
+    decidedBy: { type: 'string' },
+    decidedOn: { type: 'integer' },
+    decisionNote: { type: 'text' },
+  };
+
+  // The one request that blocks a new one: same user, same group, still open.
+  // Denied/cancelled requests deliberately do not block -- circumstances change
+  // and a user may ask again.
+  static async findOpen(uid, groupCn) {
+    const rows = await this.list({ where: { uid, groupCn, status: STATUS.PENDING } });
+    return rows[0] || null;
+  }
+
+  static async listForUser(uid) {
+    return this.list({ where: { uid } });
+  }
+
+  static async listPending() {
+    return this.list({ where: { status: STATUS.PENDING } });
+  }
+}
+
+module.exports = { AccessRequest, STATUS };

nodejs/models/agent.js

@@ -0,0 +1,185 @@
+'use strict';
+
+const crypto = require('crypto');
+const { Model } = require('@simpleworkjs/orm');
+
+// A theta-agent enrolled against this SSO.
+//
+// Before this model existed the "agent token" was generated in the browser and
+// never recorded anywhere, so the server had no way to tell an agent it issued
+// from one someone invented -- /api/agent/ws accepted any string, and there was
+// no way to revoke a token or to know that an agent existed while it was
+// offline. The row is now the authority: an agent is only real if it is here.
+//
+// The raw token is shown exactly once, at enrollment. Only its SHA-256 lands in
+// the database, so a database disclosure does not hand over working agent
+// credentials. `tokenPrefix` is the first 8 characters, kept in the clear so the
+// UI and logs can identify an agent without holding the secret.
+class Agent extends Model {
+	// Tokens are compared by hash on every WebSocket connect. SHA-256 (not
+	// bcrypt) is deliberate: this runs on the connection path and the token is a
+	// 256-bit random value, not a human-chosen password, so there is nothing for
+	// a slow KDF to protect against here.
+	static hashToken(raw) {
+		return crypto.createHash('sha256').update(String(raw || ''), 'utf8').digest('hex');
+	}
+
+	static generateToken() {
+		return crypto.randomBytes(32).toString('hex');
+	}
+
+	// Resolve a presented token to its (non-revoked) agent, or null. Every
+	// caller that authenticates an agent must go through here.
+	static async authenticate(rawToken) {
+		if (!rawToken || typeof rawToken !== 'string') return null;
+		const tokenHash = this.hashToken(rawToken);
+		const matches = await this.list({ where: { tokenHash } });
+		const agent = matches && matches[0];
+		if (!agent) return null;
+		if (agent.revoked) return null;
+		return agent;
+	}
+
+	// Enroll a new agent and return { agent, token }. The caller is responsible
+	// for showing `token` to the operator once and never storing it.
+	static async enroll({ name, resourceId, enrolledBy, description }) {
+		const token = this.generateToken();
+		const agent = await this.create({
+			id: crypto.randomUUID(),
+			name: name || 'theta-agent',
+			description: description || null,
+			tokenHash: this.hashToken(token),
+			tokenPrefix: token.slice(0, 8),
+			resourceId: resourceId || null,
+			revoked: false,
+			enrolled_by: enrolledBy || null,
+			enrolled_on: Math.floor(Date.now() / 1000)
+		});
+		return { agent, token };
+	}
+
+	// Issue a fresh token for an existing agent, invalidating the old one.
+	async rotateToken() {
+		const token = Agent.generateToken();
+		await this.update({
+			tokenHash: Agent.hashToken(token),
+			tokenPrefix: token.slice(0, 8),
+			revoked: false
+		});
+		return token;
+	}
+
+	static fields = {
+		id: { type: 'uuid', primaryKey: true },
+		name: { type: 'string', isRequired: true },
+		description: { type: 'text' },
+		// Never the raw token. See hashToken above.
+		tokenHash: { type: 'string', isRequired: true },
+		tokenPrefix: { type: 'string' },
+		// The host this agent runs on. Nullable so an agent can be enrolled
+		// before its host exists in the Directory, but the UI pushes for it:
+		// without this link there is nothing to hang resource control off, and
+		// the old code had to guess by matching hostnames to slugs.
+		resource: { type: 'hasOne', model: 'Resource' }, // creates resourceId
+		revoked: { type: 'boolean', default: false },
+		enrolled_by: { type: 'string' },
+		enrolled_on: { type: 'integer' },
+		// Survives a restart, which the in-memory map did not: an agent that is
+		// installed but currently down is now distinguishable from one that was
+		// never enrolled.
+		version: { type: 'string' },
+		last_seen: { type: 'integer' },
+		last_ip: { type: 'string' },
+		lastDiscovery: { type: 'json', default: {} },
+		lastTelemetry: { type: 'json', default: {} }
+	};
+
+	// The shape the admin API returns. Never includes tokenHash.
+	toPublic(liveState) {
+		const data = this.toJSON ? this.toJSON() : { ...this };
+		delete data.tokenHash;
+		return {
+			...data,
+			version: data.version || (data.lastDiscovery && data.lastDiscovery.version) || (data.lastTelemetry && data.lastTelemetry.version) || 'unknown',
+			lastSeen: data.last_seen ? new Date(data.last_seen * 1000).toISOString() : null,
+			connected: !!(liveState && liveState.connected),
+			// "Online" is a live-connection fact, not a stored one. A row with a
+			// last_seen from an hour ago is an installed agent that is down.
+			isOnline: !!(liveState && liveState.connected),
+			lastResponse: (liveState && liveState.lastResponse) || null
+		};
+	}
+}
+
+// A join key: the one credential an operator hands out so a host can enroll
+// itself. Requiring an admin to pre-register every machine before the agent
+// would talk to them made adding a host a two-system chore -- installing the
+// agent should be enough.
+//
+// A join key is NOT the agent's long-term credential. On first connect the
+// server auto-enrolls the host and issues it a unique per-agent token, which
+// the agent persists and uses from then on (PROTOCOL.md 1.2). That keeps the
+// operator experience to "one key" while still giving every host its own
+// revocable identity -- revoking a single agent means something, and a host
+// that is compromised does not hand over the credential for the whole fleet.
+class AgentJoinKey extends Model {
+	static hashKey(raw) {
+		return crypto.createHash('sha256').update(String(raw || ''), 'utf8').digest('hex');
+	}
+
+	static generateKey() {
+		// `tjk_` so an operator can tell a join key from an agent token at a
+		// glance -- they are handled very differently.
+		return 'tjk_' + crypto.randomBytes(32).toString('hex');
+	}
+
+	// Resolve a presented key to a usable join key, or null. Expiry and
+	// revocation are both enforced here so no caller can forget one.
+	static async authenticate(rawKey) {
+		if (!rawKey || typeof rawKey !== 'string') return null;
+		const keyHash = this.hashKey(rawKey);
+		const matches = await this.list({ where: { keyHash } });
+		const key = matches && matches[0];
+		if (!key) return null;
+		if (key.revoked) return null;
+		if (key.expires_on && key.expires_on < Math.floor(Date.now() / 1000)) return null;
+		return key;
+	}
+
+	static async issue({ label, createdBy, expiresInDays }) {
+		const raw = this.generateKey();
+		const key = await this.create({
+			id: crypto.randomUUID(),
+			label: label || 'default',
+			keyHash: this.hashKey(raw),
+			keyPrefix: raw.slice(0, 12),
+			revoked: false,
+			created_by: createdBy || null,
+			created_on: Math.floor(Date.now() / 1000),
+			expires_on: expiresInDays ? Math.floor(Date.now() / 1000) + expiresInDays * 86400 : null,
+			use_count: 0
+		});
+		return { key, raw };
+	}
+
+	static fields = {
+		id: { type: 'uuid', primaryKey: true },
+		label: { type: 'string', isRequired: true },
+		keyHash: { type: 'string', isRequired: true },
+		keyPrefix: { type: 'string' },
+		revoked: { type: 'boolean', default: false },
+		created_by: { type: 'string' },
+		created_on: { type: 'integer' },
+		expires_on: { type: 'integer' },
+		use_count: { type: 'integer', default: 0 },
+		last_used_on: { type: 'integer' }
+	};
+
+	toPublic() {
+		const data = this.toJSON ? this.toJSON() : { ...this };
+		delete data.keyHash;
+		return data;
+	}
+}
+
+module.exports = { Agent, AgentJoinKey };

nodejs/models/auth.js

@@ -23,7 +23,7 @@ Auth.login = async function(data){
 
 		return {user, token}
 	}catch(error){
-		console.error("AUTH LOGIN error:", error);
+		console.error("AUTH LOGIN error:", error.name, error.message);
 		throw this.errors.login();
 	}
 };

nodejs/models/email.js

@@ -33,8 +33,15 @@ Mail.send = function(to, subject, message, from){
 
 		var transporter = nodemailer.createTransport(transportOpts);
 
+		// Most authenticated SMTP relays (and this bit the field: "554 5.7.1
+		// ...: Sender is not same as SMTP authenticate username") require the
+		// envelope/header From to equal the authenticated user, or reject the
+		// send outright. If the operator hasn't set an explicit smtp.from,
+		// defaulting to the SMTP username is far more likely to actually send
+		// than a made-up noreply@theta42.com address that no relay authorized
+		// this account to send as.
 		var mailOpts = {
-			from: from || conf.smtp.from || `${conf.name} Accounts <noreply@theta42.com>`,
+			from: from || conf.smtp.from || conf.smtp.user || `${conf.name} Accounts <noreply@theta42.com>`,
 			to: to,
 			subject: subject,
 			html: message
@@ -58,7 +65,7 @@ Mail.sendTemplate = async function(to, template, context, from){
 		to,
 		mustache.render(template.subject, context),
 		mustache.render(template.message, context),
-		from || (template.from && mustache.render(template.message, context))
+		from || (template.from && mustache.render(template.from, context))
 	)
 };
 

nodejs/models/group_ldap.js

@@ -3,23 +3,22 @@
 const { Client, Attribute, Change } = require('ldapts');
 const { LRUCache } = require('lru-cache');
 const conf = require('@simpleworkjs/conf').ldap;
+// Connection + escaping from the shared @simpleworkjs/ldap package. Local
+// wrappers preserve the no-arg call signatures; see user_ldap.js for rationale.
+const { makeClient: _makeClient, withClient: _withClient, escapeFilter, escapeDN } = require('@simpleworkjs/ldap');
+const escapeLDAPSearchValue = escapeFilter;
+const escapeLDAPDNValue = escapeDN;
 
 function makeClient() {
-	return new Client({ url: conf.url });
+	return _makeClient(conf);
 }
 
 async function withClient(fn) {
-	const client = makeClient();
-	try {
-		await client.bind(conf.bindDN, conf.bindPassword);
-		return await fn(client);
-	} finally {
-		await client.unbind().catch(() => {});
-	}
+	return _withClient(conf, fn);
 }
 
 async function getGroups(client, member){
-	let memberFilter = member ? `(member=${member})`: ''
+	let memberFilter = member ? `(member=${escapeLDAPSearchValue(member)})`: ''
 
 	let groups = (await client.search(conf.groupBase, {
 		scope: 'sub',
@@ -35,7 +34,8 @@ async function getGroups(client, member){
 }
 
 async function addGroup(client, data){
-	await client.add(`cn=${data.name},${conf.groupBase}`, {
+	const safeName = escapeLDAPDNValue(data.name);
+	await client.add(`cn=${safeName},${conf.groupBase}`, {
 		cn: data.name,
 		member: data.owner,
 		description: data.description,
@@ -112,18 +112,190 @@ async function cachedListDetail() {
 	return promise;
 }
 
+// --- Nested groups -------------------------------------------------------
+//
+// `groupOfNames.member` holds DNs, and nothing says those DNs must be users --
+// a group DN is a perfectly legal member. That is how nesting is stored here:
+// as-is, no extra schema, no denormalization, the nesting visible in LDAP
+// exactly as an admin entered it.
+//
+// What LDAP will NOT do is resolve it. The memberof overlay records only
+// *direct* membership, and a `(member=<dn>)` filter likewise finds only the
+// groups that list the DN literally. So transitivity is computed here, and
+// every membership question in the app must go through these helpers or it
+// will silently see one level and grant nothing for a nested group.
+//
+// The whole group set is one subtree search, so the closure is computed in
+// memory rather than issuing a query per level. `resolverCache` keeps that
+// search off the hot path for bursts; it is cleared by every write below, so
+// the only staleness it can introduce is from edits made outside this app.
+// Auth decisions ride on this, hence the deliberately short TTL.
+
+const NESTING_TTL_MS = 15 * 1000;
+const MAX_NESTING_DEPTH = Number(conf.groupNestingDepth) > 0 ? Number(conf.groupNestingDepth) : 10;
+
+const resolverCache = new LRUCache({ max: 1, ttl: NESTING_TTL_MS, ttlAutopurge: true });
+
+async function allGroupsForResolver() {
+	const hit = resolverCache.get('all');
+	if (hit) return hit;
+	const promise = withClient(async (client) => {
+		const groups = await getGroups(client);
+		return groups.map(g => ({ ...g }));
+	}).then(plain => {
+		resolverCache.set('all', plain);
+		return plain;
+	}).catch(err => {
+		resolverCache.delete('all');
+		throw err;
+	});
+	resolverCache.set('all', promise);
+	return promise;
+}
+
+const lc = dn => String(dn || '').toLowerCase();
+
+// dn -> [groups that list dn as a member]. One pass, reused for every lookup.
+function buildParentIndex(groups) {
+	const parents = new Map();
+	for (const group of groups) {
+		for (const member of [].concat(group.member || []).filter(Boolean)) {
+			const key = lc(member);
+			if (!parents.has(key)) parents.set(key, []);
+			parents.get(key).push(group);
+		}
+	}
+	return parents;
+}
+
+// Every group `dn` belongs to, directly or through any chain of nested groups.
+// Breadth-first with a visited set, so a cycle (A in B, B in A) terminates
+// instead of hanging, and MAX_NESTING_DEPTH bounds a pathological chain.
+function closureUp(dn, groups) {
+	const parents = buildParentIndex(groups);
+	const found = new Map(); // cn -> group
+	const seen = new Set([lc(dn)]);
+	let frontier = [lc(dn)];
+
+	for (let depth = 0; depth < MAX_NESTING_DEPTH && frontier.length; depth++) {
+		const next = [];
+		for (const current of frontier) {
+			for (const group of parents.get(current) || []) {
+				const groupDn = lc(group.dn);
+				if (seen.has(groupDn)) continue;
+				seen.add(groupDn);
+				found.set(group.cn, group);
+				// The group itself is now a member to look up: this is the step
+				// that makes the walk transitive rather than one-level.
+				next.push(groupDn);
+			}
+		}
+		frontier = next;
+	}
+	return [...found.values()];
+}
+
+// Every member DN reachable from a group, split into the users it effectively
+// grants and the groups it nests. `direct` is kept separate so the UI can show
+// "3 members, 12 effective" and so removal stays unambiguous.
+function closureDown(group, groups) {
+	const byDn = new Map(groups.map(g => [lc(g.dn), g]));
+	const users = new Set();
+	const nested = new Map();
+	const seen = new Set([lc(group.dn)]);
+	let frontier = [group];
+
+	for (let depth = 0; depth < MAX_NESTING_DEPTH && frontier.length; depth++) {
+		const next = [];
+		for (const current of frontier) {
+			for (const member of [].concat(current.member || []).filter(Boolean)) {
+				const key = lc(member);
+				const asGroup = byDn.get(key);
+				if (asGroup) {
+					if (seen.has(key)) continue;
+					seen.add(key);
+					nested.set(asGroup.cn, asGroup);
+					next.push(asGroup);
+				} else {
+					users.add(member);
+				}
+			}
+		}
+		frontier = next;
+	}
+	return { users: [...users], nested: [...nested.values()] };
+}
+
 var Group = {};
 
+// Set when slapd carries the nestgroup overlay (docker-entrypoint.sh exports
+// app_ldap__nestedGroupsServerSide=true after detecting nestgroup.so). With it,
+// a plain `(member=<dn>)` search already returns the full transitive set and the
+// in-app closure is redundant work on every request. Without it -- e.g. pointed
+// at a stock 2.6.x server, which no release ships nestgroup in -- the app must
+// compute the closure itself or nested groups silently grant nothing.
+const SERVER_SIDE_NESTING = String(conf.nestedGroupsServerSide) === 'true';
+
+// Transitive: every group CN this member belongs to, at any nesting depth.
+// Callers making an access decision must use this rather than reading
+// `memberOf`, which a server without nestgroup only ever populates one level
+// deep.
 Group.list = async function(member){
 	if (member) {
-		return withClient(async (client) => {
-			const groups = await getGroups(client, member);
-			return groups.map(group => group.cn);
-		});
+		if (SERVER_SIDE_NESTING) {
+			return withClient(async (client) => {
+				const groups = await getGroups(client, member);
+				return groups.map(group => group.cn);
+			});
+		}
+		const groups = await allGroupsForResolver();
+		return closureUp(member, groups).map(group => group.cn);
 	}
 	return (await cachedListDetail()).map(group => group.cn);
 }
 
+// The members a group effectively grants: users reached through any chain of
+// nested groups, plus the nested groups themselves for display.
+Group.effectiveMembers = async function(cn){
+	const groups = await allGroupsForResolver();
+	const group = groups.find(g => g.cn === cn);
+	if (!group) {
+		let error = new Error('GroupNotFound');
+		error.name = 'GroupNotFound';
+		error.message = `LDAP:${cn} does not exists`;
+		error.status = 404;
+		throw error;
+	}
+	const { users, nested } = closureDown(group, groups);
+	const directMembers = [].concat(group.member || []).filter(Boolean);
+	const groupDns = new Set(groups.map(g => lc(g.dn)));
+	return {
+		cn: group.cn,
+		direct: directMembers.filter(dn => !groupDns.has(lc(dn))),
+		nestedGroups: nested.map(g => ({ cn: g.cn, dn: g.dn })),
+		effective: users,
+	};
+};
+
+// Would adding `childDn` to `parentCn` create a cycle? A group may not contain
+// itself, nor anything that already (transitively) contains it -- such a chain
+// makes membership unanswerable, and callers would rely on the depth cap to
+// stop rather than getting a real answer.
+Group.wouldCycle = async function(parentCn, childDn){
+	const groups = await allGroupsForResolver();
+	const parent = groups.find(g => g.cn === parentCn);
+	if (!parent) return false;
+	if (lc(parent.dn) === lc(childDn)) return true;
+	const child = groups.find(g => lc(g.dn) === lc(childDn));
+	if (!child) return false; // a user DN can never close a cycle
+	// Adding child under parent is a cycle exactly when parent is already
+	// reachable downward from child.
+	const { nested } = closureDown(child, groups);
+	return nested.some(g => lc(g.dn) === lc(parent.dn));
+};
+
+Group.clearResolverCache = function(){ resolverCache.clear(); };
+
 Group.listDetail = async function(member){
 	if (member) {
 		return withClient(async (client) => getGroups(client, member));
@@ -139,9 +311,10 @@ Group.get = async function(data){
 	}
 
 	return withClient(async (client) => {
+		const safeName = escapeLDAPSearchValue(data.name);
 		let group = (await client.search(conf.groupBase, {
 			scope: 'sub',
-			filter: `(&(objectClass=groupOfNames)(cn=${data.name}))`,
+			filter: `(&(objectClass=groupOfNames)(cn=${safeName}))`,
 			attributes: ['cn', 'description', 'member', 'owner', 'createTimestamp', 'modifyTimestamp'],
 		})).searchEntries[0];
 
@@ -165,6 +338,7 @@ Group.add = async function(data){
 	return withClient(async (client) => {
 		await addGroup(client, data);
 		cache.clear();
+		resolverCache.clear();
 		return this.get(data);
 	});
 }
@@ -173,6 +347,7 @@ Group.addMember = async function(user){
 	await withClient(async (client) => addMember(client, this, user));
 	this.member = [].concat(this.member || []).concat([user.dn]);
 	cache.clear();
+	resolverCache.clear();
 	return this;
 };
 
@@ -185,6 +360,7 @@ Group.removeMember = async function(user){
 	}
 	this.member = [].concat(this.member || []).filter(dn => dn !== user.dn);
 	cache.clear();
+	resolverCache.clear();
 	return this;
 };
 
@@ -192,6 +368,7 @@ Group.addOwner = async function(user){
 	await withClient(async (client) => addOwner(client, this, user));
 	this.owner = [].concat(this.owner || []).concat([user.dn]);
 	cache.clear();
+	resolverCache.clear();
 	return this;
 };
 
@@ -204,12 +381,14 @@ Group.removeOwner = async function(user){
 	}
 	this.owner = [].concat(this.owner || []).filter(dn => dn !== user.dn);
 	cache.clear();
+	resolverCache.clear();
 	return this;
 };
 
 Group.remove = async function(){
 	await withClient(async (client) => client.del(this.dn));
 	cache.clear();
+	resolverCache.clear();
 	return true;
 }
 

nodejs/models/index.js

@@ -1,14 +1,84 @@
 'use strict';
 
 const conf = require('@simpleworkjs/conf');
-const {setUpTable} = require('model-redis');
+const { setUpTable } = require('model-redis');
 
+// Keep model-redis for the ones not yet ported
 const Table = setUpTable(conf.redis);
-
 module.exports = Table;
 
-require('./token');
+const { Token, AuthToken, InviteToken, ImpersonationToken, PasswordResetToken, OtpToken, ServiceToken } = require('./token');
 require('./verification');
-require('./oauth_client');
 require('./oauth_code');
 require('./api_token');
+
+const { init } = require('@simpleworkjs/orm');
+const { Resource, ResourceEdge, ResourceGroup } = require('./resource');
+const { AccessRequest } = require('./access_request');
+const { Webhook } = require('./webhook');
+const { PluginInstance } = require('./plugin_instance');
+const { SharedSecret } = require('./shared_secret');
+const { SharedSecretGrant } = require('./shared_secret_grant');
+const { VaultAppToken } = require('./vault_app_token');
+const { Agent, AgentJoinKey } = require('./agent');
+const { SiteJoinKey } = require('./site_join_key');
+const { SiteSpoke } = require('./site_spoke');
+async function initORM() {
+  const ormConf = conf.orm || {
+    dialect: 'sqlite',
+    storage: './config/inventory.sqlite',
+    logging: false
+  };
+  ormConf.redis = conf.redis;
+
+  console.log('[initORM] Starting ORM initialization...');
+  try {
+    await init({
+      conf: { orm: ormConf },
+      models: [
+        Resource, ResourceEdge, ResourceGroup, AccessRequest, Webhook, PluginInstance,
+        SharedSecret, SharedSecretGrant, VaultAppToken, Agent, AgentJoinKey, SiteJoinKey, SiteSpoke,
+        Token, AuthToken, InviteToken, ImpersonationToken, PasswordResetToken, OtpToken, ServiceToken
+      ]
+    });
+    console.log('[initORM] ORM initialized successfully');
+    console.log('[initORM] Resource.orm =', !!Resource.orm, 'Token.orm =', !!Token.orm);
+    await healSchema();
+  } catch (err) {
+    console.error('[initORM] ORM initialization failed:', err.message);
+    throw err;
+  }
+}
+
+// Add-only schema heal. @simpleworkjs/orm runs sequelize.sync() WITHOUT alter,
+// which creates missing tables but never touches existing ones — so a column
+// added in a newer release (e.g. PluginInstance.lastLog) simply never appears
+// in an upgraded deployment's database and every query on the model fails
+// ("no such column"). This walks each Sequelize model and ADDs any attribute
+// missing from its table. Strictly additive (never drops or retypes), works on
+// any dialect via the query interface, and fail-soft per column so one bad
+// attribute can't take the boot down.
+async function healSchema() {
+  const adapter = Resource.orm && Resource.orm.adapters && Resource.orm.adapters.sequelize;
+  if (!adapter || !adapter.sequelize) return;
+  const sequelize = adapter.sequelize;
+  const qi = sequelize.getQueryInterface();
+  for (const SM of Object.values(sequelize.models)) {
+    const table = SM.getTableName();
+    let existing;
+    try { existing = await qi.describeTable(table); }
+    catch (e) { continue; } // no table yet — sync() handles creation
+    for (const [name, attr] of Object.entries(SM.getAttributes())) {
+      const col = attr.field || name;
+      if (existing[col]) continue;
+      try {
+        await qi.addColumn(table, col, attr);
+        console.log(`[initORM] schema heal: added missing column ${table}.${col}`);
+      } catch (e) {
+        console.error(`[initORM] schema heal: could not add ${table}.${col}:`, e.message);
+      }
+    }
+  }
+}
+
+module.exports.initORM = initORM;

nodejs/models/oauth_client.js

@@ -1,50 +1,137 @@
 'use strict';
 
-const Table = require('.');
+const { Resource } = require('./resource');
 const bcrypt = require('bcrypt');
-const UUID = function b(a){return a?(a^Math.random()*16>>a/4).toString(16):([1e7]+-1e3+-4e3+-8e3+-1e11).replace(/[018]/g,b)};
+const crypto = require('crypto');
 const conf = require('@simpleworkjs/conf');
+const UUID = () => crypto.randomUUID();
 
 const defaultLifetime = (conf.oauth && conf.oauth.token_lifetime) || {
 	access_token: 3600,
 	refresh_token: 2592000
 };
 
-class OAuthClient extends Table {
-	static _key = 'client_id';
-	static _keyMap = {
-		'client_id':          {default: UUID, type: 'string'},
-		'client_secret_hash': {isRequired: true, type: 'string', isPrivate: true},
-		'name':               {isRequired: true, type: 'string', min: 1, max: 255},
-		'description':        {default: '', type: 'string'},
-		'redirect_uris':      {default: [], type: 'object'},
-		'scopes':             {default: ['openid', 'profile', 'email', 'groups'], type: 'object'},
-		'allowed_groups':     {default: [], type: 'object'},
-		'token_lifetime':     {default: function(){ return Object.assign({}, defaultLifetime) }, type: 'object'},
-		'created_by':         {isRequired: true, type: 'string'},
-		'created_on':         {default: function(){ return (new Date).getTime() }},
-		'is_valid':           {default: true, type: 'boolean'},
-	}
-
+class OAuthClient {
 	static async add(data) {
-		const raw_secret = UUID();
-		data.client_secret_hash = await bcrypt.hash(raw_secret, 10);
-		data.client_id = UUID();
-		const client = await this.create(data);
-		client._raw_secret = raw_secret;
-		return client;
+		const raw_secret = crypto.randomUUID();
+		const client_id = crypto.randomUUID();
+		const client_secret_hash = await bcrypt.hash(raw_secret, 10);
+
+		// Generate a unique slug from the client name
+		let slug = data.name.toLowerCase().replace(/[^a-z0-9]+/g, '-').replace(/^-|-$/g, '') || 'oauth-client';
+		// Ensure uniqueness by appending a suffix if needed
+		const existing = await Resource.list({ where: { slug } });
+		if (existing.length) slug = `${slug}-${client_id.slice(0, 8)}`;
+
+		const r = await Resource.create({
+			id: client_id,
+			kind: 'oauth',
+			name: data.name,
+			slug: slug,
+			description: data.description || '',
+			owner: data.created_by,
+			metadata: {
+				client_secret_hash,
+				redirect_uris: data.redirect_uris || [],
+				scopes: data.scopes || ['openid', 'profile', 'email', 'groups'],
+				allowed_groups: data.allowed_groups || [],
+				token_lifetime: data.token_lifetime || { ...defaultLifetime }
+			}
+		});
+		
+		r._raw_secret = raw_secret;
+		r.client_id = client_id;
+		return r;
+	}
+	static async get(client_id) {
+			const notFound = () => {
+				const e = new Error('OAuthClient not found');
+				e.status = 404;
+				return e;
+			};
+			let r;
+			try {
+				r = await Resource.get(client_id);
+			} catch (_) {
+				throw notFound();
+			}
+			// Resource.get() returns null (does not throw) for a missing id —
+			// guard it so a bad/undefined client_id is a clean 404, not a
+			// "Cannot read properties of null (reading 'kind')" 500.
+			if (!r || r.kind !== 'oauth') throw notFound();
+		// Map metadata to top-level properties to satisfy routes/oauth.js without rewriting it
+		r.client_id = r.id;
+		r.client_secret_hash = r.metadata.client_secret_hash;
+		r.redirect_uris = r.metadata.redirect_uris || [];
+		r.scopes = r.metadata.scopes || ['openid', 'profile', 'email', 'groups'];
+		r.allowed_groups = r.metadata.allowed_groups || [];
+		r.token_lifetime = r.metadata.token_lifetime || { ...defaultLifetime };
+		// Resource has no is_valid column; validity lives in metadata (absent = valid)
+		r.is_valid = r.metadata.is_valid !== false;
+		r.verifySecret = async (secret) => bcrypt.compare(secret, r.client_secret_hash);
+		
+		r.rotateSecret = async () => {
+			const raw_secret = crypto.randomUUID();
+			r.metadata.client_secret_hash = await bcrypt.hash(raw_secret, 10);
+			await r.update({ metadata: r.metadata });
+			return raw_secret;
+		};
+
+		// The ORM Model.toJSON() only serializes schema fields, so the mapped
+		// properties above (client_id, scopes, redirect_uris, …) would be
+		// stripped from any res.json() — that's why GET /api/oauth/client
+		// returned client_id: undefined and the bootstrap's rotate blew up.
+		// Emit the public shape explicitly. client_secret_hash is deliberately
+		// omitted so it never leaks over the API.
+		r.toJSON = function () {
+			return {
+				client_id: r.id,
+				id: r.id,
+				kind: r.kind,
+				name: r.name,
+				slug: r.slug,
+				owner: r.owner,
+				description: r.description,
+				redirect_uris: r.redirect_uris,
+				scopes: r.scopes,
+				allowed_groups: r.allowed_groups,
+				token_lifetime: r.token_lifetime,
+				is_valid: r.is_valid,
+			};
+		};
+
+		// proxy update to handle metadata correctly
+		const originalUpdate = r.update.bind(r);
+		r.update = async (data) => {
+			if (data.redirect_uris !== undefined) r.metadata.redirect_uris = data.redirect_uris;
+			if (data.scopes !== undefined) r.metadata.scopes = data.scopes;
+			if (data.allowed_groups !== undefined) r.metadata.allowed_groups = data.allowed_groups;
+			if (data.token_lifetime !== undefined) r.metadata.token_lifetime = data.token_lifetime;
+			if (data.is_valid !== undefined) r.metadata.is_valid = data.is_valid;
+
+			const updateData = { metadata: r.metadata };
+			if (data.name !== undefined) updateData.name = data.name;
+			if (data.description !== undefined) updateData.description = data.description;
+			
+			return originalUpdate(updateData);
+		};
+
+		return r;
 	}
 
-	async verifySecret(secret) {
-		return bcrypt.compare(secret, this.client_secret_hash);
+	static async list() {
+		const resources = await Resource.list({ where: { kind: 'oauth' } });
+		return Promise.all(resources.map(r => this.get(r.id)));
 	}
 
-	async rotateSecret() {
-		const raw_secret = UUID();
-		await this.update({ client_secret_hash: await bcrypt.hash(raw_secret, 10) });
-		return raw_secret;
+	static async listDetail() {
+		return this.list();
+	}
+
+	static async verifySecret(client_id, secret) {
+		const client = await this.get(client_id);
+		return client.verifySecret(secret);
 	}
 }
-OAuthClient.register();
 
 module.exports = { OAuthClient };

nodejs/models/oauth_code.js

@@ -1,7 +1,8 @@
 'use strict';
 
 const Table = require('.');
-const UUID = function b(a){return a?(a^Math.random()*16>>a/4).toString(16):([1e7]+-1e3+-4e3+-8e3+-1e11).replace(/[018]/g,b)};
+const crypto = require('crypto');
+const UUID = () => crypto.randomUUID();
 
 // Shared base keyMap matching Token's schema so these behave as tokens
 const tokenKeyMap = {

nodejs/models/plugin_instance.js

@@ -0,0 +1,83 @@
+'use strict';
+
+// PluginInstance — the registry of configured, loadable plugin copies.
+//
+// The SSO plugin system (see nodejs/services/plugin_registry.js) distinguishes
+// **plugin types** (the .js modules under nodejs/plugins/<category>/<type>.js)
+// from **plugin instances** — a configured, loadable/unloadable *copy* of a
+// type. You can have several instances of the same type (e.g. two Proxmox
+// endpoints with their own URLs + tokens), each on its own schedule.
+//
+// This table holds the *non-secret* per-instance state: which type it is, its
+// schedule (cron), whether it's loaded (enabled), and its non-secret config.
+// Per-instance **secrets** (the configSchema fields flagged `secret:true`,
+// e.g. a Proxmox `tokenSecret` or UniFi `password`) live in OpenBao at
+// `secret/plugins/<id>/conf` (see nodejs/utils/plugin_secrets.js) — never in
+// the DB. The DB row's `config` JSON column holds only non-secret field values.
+//
+// `slug` is the discovery source name passed to DiscoveryReconciler.reconcile,
+// so a discovery instance's resources are attributed to a stable, human-chosen
+// name rather than its uuid. Unique, so two instances can't shadow each other
+// in the resource graph's `discovery_sources`.
+//
+// Like Resource/AccessRequest, there is no ORM auto-timestamp hook: the route
+// handler stamps created_by/on + updated_by/on explicitly on every write (see
+// routes/api_plugins.js). `id` (uuid) is generated by the ORM on create.
+
+const { Model } = require('@simpleworkjs/orm');
+
+const STATUS = {
+  OK: 'ok',
+  ERROR: 'error',
+  RUNNING: 'running',
+};
+
+class PluginInstance extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    // A registered plugin type slug (matches a manifest `type`). Validated
+    // against the registry before a row is created.
+    pluginType: { type: 'string', isRequired: true, min: 1, max: 64 },
+    // The plugin's category (e.g. 'discovery'). Copied from the manifest at
+    // create time so the scheduler can dispatch without re-reading the registry
+    // on every run (and so a later type removal still shows what the instance was).
+    category: { type: 'string', isRequired: true, default: 'discovery', min: 1, max: 64 },
+    // Human label for the instance.
+    name: { type: 'string', isRequired: true, min: 1, max: 120 },
+    // Stable handle: discovery source name + unique constraint. Lowercase
+    // alnum + hyphen/underscore to stay safe as a resource-graph slug.
+    slug: { type: 'string', isRequired: true, unique: true, min: 1, max: 64 },
+    // Loaded into the scheduler? `false` = unloaded (no scheduled runs).
+    enabled: { type: 'boolean', default: true },
+    // Cron schedule (5-field). The scheduler turns this into a BullMQ
+    // repeatable JobScheduler.
+    cron: { type: 'string', isRequired: true, default: '0 * * * *' },
+    // Non-secret configSchema field values. Secret fields are NOT here.
+    config: { type: 'json', default: {} },
+    // Last-run bookkeeping, updated by the scheduler worker.
+    lastRunAt: { type: 'integer' },
+    lastStatus: { type: 'string' },
+    lastError: { type: 'text' },
+    lastLog: { type: 'text' },
+    // Audit stamps (set by the route handler, not by an ORM hook).
+    created_by: { type: 'string' },
+    created_on: { type: 'integer' },
+    updated_by: { type: 'string' },
+    updated_on: { type: 'integer' },
+  };
+
+  // All instances the scheduler should run: enabled only. Loaded fresh each
+  // boot / load; not cached on the model (the scheduler is the source of truth
+  // for what's actually scheduled).
+  static async listEnabled() {
+    return this.list({ where: { enabled: true } });
+  }
+
+  // Look up by slug — used by tests + the reconciler when only a slug is known.
+  static async getBySlug(slug) {
+    const rows = await this.list({ where: { slug } });
+    return rows[0] || null;
+  }
+}
+
+module.exports = { PluginInstance, STATUS };

nodejs/models/resource.js

@@ -0,0 +1,250 @@
+const crypto = require('crypto');
+const { Model } = require('@simpleworkjs/orm');
+
+const { Group } = require('./group_ldap');
+
+class Resource extends Model {
+  static exposedMethods = [
+    { method: 'search', route: 'resources', verb: 'get', args: { from: 'query' } },
+    { method: 'getBySlug', route: 'resources/:slug', verb: 'get', args: { from: 'params', names: ['slug'] } },
+    { method: 'getGraph', route: 'graph', verb: 'get' },
+    { method: 'getMyAccess', route: 'me', verb: 'get', args: { from: 'user' } }
+  ];
+
+  static async search(query) {
+    const graph = await this.getGraph();
+    let resources = graph.resources;
+    
+    if (query.kind) {
+      resources = resources.filter(r => r.kind === query.kind);
+    }
+    
+    if (query.group) {
+      const rgs = await ResourceGroup.list({ where: { groupCn: query.group } });
+      const allowedIds = new Set(rgs.map(rg => rg.resourceId));
+      resources = resources.filter(r => allowedIds.has(r.id));
+    }
+
+    if (query.parent) {
+      const parents = graph.resources.filter(r => r.slug === query.parent);
+      if (parents.length > 0) {
+        const parentId = parents[0].id;
+        const childIds = new Set(graph.edges.filter(e => e.parentId === parentId).map(e => e.childId));
+        resources = resources.filter(r => childIds.has(r.id));
+      } else {
+        resources = [];
+      }
+    }
+    return resources;
+  }
+
+  static async getBySlug(slug) {
+    const graph = await this.getGraph();
+    const resource = graph.resources.find(r => r.slug === slug);
+    if (!resource) {
+      let err = new Error('Resource not found');
+      err.status = 404;
+      throw err;
+    }
+    
+    const parents = graph.edges.filter(e => e.childId === resource.id);
+    const children = graph.edges.filter(e => e.parentId === resource.id);
+    
+    return {
+      ...resource,
+      parents,
+      children
+    };
+  }
+
+  static async getGraph() {
+    const resources = await this.list();
+    const edges = await ResourceEdge.list();
+    
+    // Convert to simple objects so we can mutate metadata properties safely
+    const resObjs = resources.map(r => {
+      const obj = r.toJSON ? r.toJSON() : { ...r };
+      obj.metadata = obj.metadata || {};
+      return obj;
+    });
+    
+    // Bubble up production status: if any child is prod, parent is prod
+    const isProdCache = new Map();
+    function checkProd(resId, visited = new Set()) {
+      if (isProdCache.has(resId)) return isProdCache.get(resId);
+      if (visited.has(resId)) return false; // Cycle prevention
+      
+      visited.add(resId);
+      const r = resObjs.find(x => x.id === resId);
+      if (!r) return false;
+      
+      // If intrinsically prod, return true
+      if (r.metadata.isProduction) {
+        isProdCache.set(resId, true);
+        return true;
+      }
+      
+      // Check children
+      const childrenIds = edges.filter(e => e.parentId === resId).map(e => e.childId);
+      for (const cid of childrenIds) {
+        if (checkProd(cid, visited)) {
+          isProdCache.set(resId, true);
+          return true;
+        }
+      }
+      
+      isProdCache.set(resId, false);
+      return false;
+    }
+    
+    let maxUpdated = 0;
+    resObjs.forEach(r => {
+      r.metadata.isProduction = checkProd(r.id);
+      if (r.updated_on && r.updated_on > maxUpdated) maxUpdated = r.updated_on;
+    });
+
+    return { resources: resObjs, edges, updated_on: maxUpdated || Date.now() };
+  }
+
+  // Stamp `resolvedAddress` on each resource: its own address/ip if it has one,
+  // otherwise the nearest ancestor's. A service usually carries no address of
+  // its own -- it is reached at the host it runs on -- so "how do I reach this"
+  // is only answerable from the graph, never from the row alone. Every caller
+  // that answers that question for a user (getMyAccess, GET /api/discovery/me)
+  // must go through here, or services come back unreachable.
+  static async withResolvedAddress(resources) {
+    if (!resources || !resources.length) return [];
+    const graph = await this.getGraph();
+
+    const resolve = (resId, visited = new Set()) => {
+      if (visited.has(resId)) return null; // prevent cycles
+      visited.add(resId);
+
+      const res = graph.resources.find(r => r.id === resId);
+      if (!res) return null;
+      if (res.metadata && res.metadata.address) return res.metadata.address;
+      if (res.metadata && res.metadata.ip) return res.metadata.ip;
+
+      for (const edge of graph.edges.filter(e => e.childId === resId)) {
+        const found = resolve(edge.parentId, visited);
+        if (found) return found;
+      }
+      return null;
+    };
+
+    return resources.map(r => {
+      const data = r.toJSON ? r.toJSON() : { ...r };
+      data.metadata = data.metadata || {};
+      data.resolvedAddress = resolve(data.id);
+      return data;
+    });
+  }
+
+  static async getMyAccess(userDn) {
+    const userGroups = await Group.list(userDn);
+    if (!userGroups || userGroups.length === 0) return [];
+
+    const resourceGroups = await ResourceGroup.list({
+      where: { groupCn: { in: userGroups } }
+    });
+
+    const resourceIds = [...new Set(resourceGroups.map(rg => rg.resourceId))];
+    if (resourceIds.length === 0) return [];
+
+    return this.withResolvedAddress(await this.list({ where: { id: { in: resourceIds } } }));
+  }
+
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    kind: { type: 'string', isRequired: true },
+    name: { type: 'string', isRequired: true },
+    slug: { type: 'string', isRequired: true, unique: true },
+    owner: { type: 'string' },
+    description: { type: 'text' },
+    metadata: { type: 'json', default: {} },
+    // Not isRequired: @simpleworkjs/orm has no auto-timestamp hook, so these
+    // are set explicitly by the route handler on every create/update (see
+    // routes/api_directory_admin.js). Existing rows predating this change
+    // simply read back undefined -- callers must render a fallback.
+    created_by: { type: 'string' },
+    created_on: { type: 'integer' },
+    updated_by: { type: 'string' },
+    updated_on: { type: 'integer' },
+    edgesAsParent: { type: 'hasMany', model: 'ResourceEdge', remoteKey: 'parentId' },
+    edgesAsChild: { type: 'hasMany', model: 'ResourceEdge', remoteKey: 'childId' },
+    groups: { type: 'hasMany', model: 'ResourceGroup', remoteKey: 'resourceId' }
+  };
+
+  // Walk parent ResourceEdges from resourceId up to the nearest ancestor
+  // whose kind === 'site', returning its slug (or null if none exists -- a
+  // top-level resource with no site parent keeps its unprefixed group name).
+  static async findAncestorSiteSlug(resourceId, visited = new Set()) {
+    if (visited.has(resourceId)) return null;
+    visited.add(resourceId);
+
+    const parentEdges = await ResourceEdge.list({ where: { childId: resourceId } });
+    for (const edge of parentEdges) {
+      const parent = await this.get(edge.parentId);
+      if (!parent) continue;
+      if (parent.kind === 'site') return parent.slug;
+      const found = await this.findAncestorSiteSlug(parent.id, visited);
+      if (found) return found;
+    }
+    return null;
+  }
+
+  // Walk all parent ResourceEdges upwards recursively to find all ancestor
+  // resources (Host, Cluster, Site, etc.).
+  static async findAllAncestors(resourceId, visited = new Set()) {
+    if (!resourceId || visited.has(resourceId)) return [];
+    visited.add(resourceId);
+
+    const ancestors = [];
+    const allEdges = await ResourceEdge.list().catch(() => []);
+    const parentEdges = allEdges.filter(e => e.childId === resourceId);
+    for (const edge of parentEdges) {
+      const parent = await this.get(edge.parentId).catch(() => null);
+      if (!parent) continue;
+      ancestors.push(parent);
+      const higher = await this.findAllAncestors(parent.id, visited);
+      ancestors.push(...higher);
+    }
+    return ancestors;
+  }
+}
+
+class ResourceEdge extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    parent: { type: 'hasOne', model: 'Resource' }, // Creates parentId
+    child: { type: 'hasOne', model: 'Resource' }, // Creates childId
+    relation: { type: 'string', isRequired: true }
+  };
+}
+
+class ResourceGroup extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    resource: { type: 'hasOne', model: 'Resource' }, // Creates resourceId
+    groupCn: { type: 'string', isRequired: true },
+    accessLevel: { type: 'string', isRequired: true }
+  };
+
+  // No DB-level unique constraint on (resourceId, groupCn) exists, so callers
+  // MUST check-then-create rather than relying on a constraint violation to
+  // catch a dupe. A caller that skips this (raw ResourceGroup.create()) and
+  // runs more than once for the same resource -- e.g. discovery reconciling
+  // the same LXC from multiple Proxmox cluster nodes -- silently accumulates
+  // duplicate access/admin rows every pass, with no error to notice it by.
+  static async ensure(resourceId, groupCn, accessLevel) {
+    const existing = await this.list({ where: { resourceId, groupCn } });
+    if (existing.length) return existing[0];
+    return this.create({ id: crypto.randomUUID(), resourceId, groupCn, accessLevel });
+  }
+}
+
+module.exports = {
+  Resource,
+  ResourceEdge,
+  ResourceGroup
+};

nodejs/models/service_account.js

@@ -1,114 +0,0 @@
-'use strict';
-
-// Non-person "service" accounts under ou=people -- bind-only LDAP identities
-// for things like theta-env's bootstrap-created cn=ldapclient (the proxy's
-// direct-LDAP bind account) or any other app/host that needs its own
-// dedicated read-only credential, as opposed to a real user who logs into
-// the web UI.
-//
-// Deliberately NOT posixAccount/inetOrgPerson (the User model's shape) --
-// these can't log into the SSO Manager UI or get a home directory/uidNumber.
-// objectClass matches exactly what theta-env's bootstrap.js already creates
-// for cn=ldapclient, so this model recognizes and manages that account too,
-// not just ones created through this UI.
-
-const { Client, Attribute, Change } = require('ldapts');
-const crypto = require('crypto');
-const conf = require('@simpleworkjs/conf').ldap;
-
-function hashPasswordSSHA512(password) {
-	const salt = crypto.randomBytes(8);
-	const hash = crypto.createHash('sha512').update(password).update(salt).digest();
-	return '{SSHA512}' + Buffer.concat([hash, salt]).toString('base64');
-}
-
-function makeClient() {
-	return new Client({ url: conf.url });
-}
-
-async function withClient(fn) {
-	const client = makeClient();
-	try {
-		await client.bind(conf.bindDN, conf.bindPassword);
-		return await fn(client);
-	} finally {
-		await client.unbind().catch(() => {});
-	}
-}
-
-const FILTER = '(&(objectClass=organizationalRole)(objectClass=simpleSecurityObject))';
-const CN_RE = /^[A-Za-z][A-Za-z0-9._-]{1,63}$/;
-
-var ServiceAccount = {};
-
-ServiceAccount.list = async function(){
-	return withClient(async (client) => {
-		const res = await client.search(conf.userBase, {
-			scope: 'sub',
-			filter: FILTER,
-			attributes: ['cn', 'description', 'createTimestamp', 'modifyTimestamp'],
-		});
-		return res.searchEntries.map((entry) => ({
-			cn: entry.cn,
-			dn: `cn=${entry.cn},${conf.userBase}`,
-			description: entry.description || '',
-			created_on: entry.createTimestamp || null,
-			modified_on: entry.modifyTimestamp || null,
-		})).sort((a, b) => a.cn.localeCompare(b.cn));
-	});
-};
-
-ServiceAccount.create = async function({cn, description}){
-	if(!cn || !CN_RE.test(cn)){
-		throw Object.assign(new Error('InvalidName'), {status: 400, message: 'Name must start with a letter and contain only letters, numbers, dot, dash, underscore.'});
-	}
-
-	const dn = `cn=${cn},${conf.userBase}`;
-	const password = crypto.randomBytes(24).toString('base64url');
-
-	await withClient(async (client) => {
-		let existing = true;
-		try{
-			const res = await client.search(dn, {scope: 'base', filter: '(objectClass=*)', attributes: ['dn']});
-			existing = res.searchEntries.length > 0;
-		}catch(error){ existing = false; }
-		if(existing){
-			throw Object.assign(new Error('NameInUse'), {status: 409, message: `"${cn}" already exists under ${conf.userBase}.`});
-		}
-
-		await client.add(dn, {
-			objectClass: ['organizationalRole', 'simpleSecurityObject', 'top'],
-			cn,
-			description: description || '',
-			userPassword: hashPasswordSSHA512(password),
-		});
-	});
-
-	return {cn, dn, description: description || '', password};
-};
-
-ServiceAccount.setPassword = async function(cn, password){
-	const dn = `cn=${cn},${conf.userBase}`;
-	const newPassword = password || crypto.randomBytes(24).toString('base64url');
-
-	await withClient(async (client) => {
-		await client.modify(dn, [
-			new Change({
-				operation: 'replace',
-				modification: new Attribute({type: 'userPassword', values: [hashPasswordSSHA512(newPassword)]}),
-			}),
-		]);
-	});
-
-	return {cn, dn, password: newPassword};
-};
-
-ServiceAccount.remove = async function(cn){
-	const dn = `cn=${cn},${conf.userBase}`;
-	await withClient(async (client) => {
-		await client.del(dn);
-	});
-	return true;
-};
-
-module.exports = {ServiceAccount};

nodejs/models/shared_secret.js

@@ -0,0 +1,56 @@
+'use strict';
+
+// SharedSecret — a secret the owner has published to the shared namespace so it
+// can be shared with other users and/or downstream apps.
+//
+// The secret DATA lives in OpenBao at `secret/shared/<ownerUid>/<slug>` (KV-v2),
+// never in the DB. This row is metadata only (owner + slug + description) and is
+// the source of truth for the UI (which shares exist). ACCESS CONTROL is enforced
+// entirely by OpenBao ACL policies: the owner's `user-<uid>` policy grants full
+// R/W on `secret/shared/<ownerUid>/*`, and each grantee's policy content is
+// edited to add `read` on the exact shared path (see vault_broker.js — policy
+// content is parsed live at token use, so a grant takes effect immediately with
+// no token re-mint). `secretId` on SharedSecretGrant links grantees to this row.
+//
+// `slug` is unique and immutable in practice — it is embedded in the shared path
+// and in grantee policy rules, so changing it would require rewriting policies.
+// Like PluginInstance, there is no ORM auto-timestamp hook: route handlers stamp
+// created_by/on + updated_by/on on every write. `id` (uuid) is generated by the
+// ORM on create.
+
+const { Model } = require('@simpleworkjs/orm');
+
+class SharedSecret extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    // Human slug embedded in the OpenBao path: secret/shared/<ownerUid>/<slug>.
+    // Unique so two owners can't collide on the same shared path.
+    slug: { type: 'string', isRequired: true, unique: true, min: 1, max: 64 },
+    // The publishing user's uid — also the shared path's namespace segment.
+    ownerUid: { type: 'string', isRequired: true, min: 1, max: 64 },
+    // Optional human description shown in the Shared tab.
+    description: { type: 'text' },
+    // Audit stamps (set by the route handler, not by an ORM hook).
+    created_by: { type: 'string' },
+    created_on: { type: 'integer' },
+    updated_by: { type: 'string' },
+    updated_on: { type: 'integer' },
+  };
+
+  // Full OpenBao KV-v2 path for this shared secret (logical path, no data/metadata).
+  static pathFor(ownerUid, slug) {
+    return `shared/${ownerUid}/${slug}`;
+  }
+
+  path() {
+    return SharedSecret.pathFor(this.ownerUid, this.slug);
+  }
+
+  // Look up by slug (unique). Returns the row or null.
+  static async getBySlug(slug) {
+    const rows = await this.list({ where: { slug } });
+    return rows[0] || null;
+  }
+}
+
+module.exports = { SharedSecret };

nodejs/models/shared_secret_grant.js

@@ -0,0 +1,53 @@
+'use strict';
+
+// SharedSecretGrant — who can read a shared secret. Each row says "grantee
+// <granteeId> (a user uid or an app name) has <capability> on the shared secret
+// <secretId>".
+//
+// This table is the metadata/UX record of a grant. The actual ENFORCEMENT lives
+// in OpenBao ACL policy content: when a grant is created, vault_broker.js
+// recomputes the grantee's policy HCL (`user-<uid>` or `app-<name>`) to include
+// `read` on the exact shared path and rewrites it. Because OpenBao parses policy
+// content live at token use, the grant applies to the grantee's existing token
+// immediately (no re-mint). Revoking removes the rule and rewrites the policy.
+//
+// granteeType distinguishes the two principal kinds:
+//   'user' — a user uid → grantee's `user-<uid>` policy is edited
+//   'app'  — an app name → grantee's `app-<name>` policy is edited (downstream apps)
+// capability is currently always 'read' (grantees are read-only); the column is
+// a string so later capabilities could be added without a migration.
+//
+// No ORM auto-timestamp hook: route handlers stamp created_by/on + updated_by/on.
+// Uniqueness on (secretId, granteeType, granteeId) prevents duplicate grants.
+
+const { Model } = require('@simpleworkjs/orm');
+
+const GRANTEE_TYPES = ['user', 'app'];
+const CAPABILITIES = ['read'];
+
+class SharedSecretGrant extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    // FK to SharedSecret.id.
+    secretId: { type: 'string', isRequired: true, min: 1 },
+    // 'user' (a uid) or 'app' (an app name) — which policy to edit.
+    granteeType: { type: 'string', isRequired: true, min: 1 },
+    // The grantee's uid (for 'user') or app name (for 'app').
+    granteeId: { type: 'string', isRequired: true, min: 1, max: 64 },
+    // Access level — 'read' today.
+    capability: { type: 'string', isRequired: true, default: 'read' },
+    // Audit stamps (set by the route handler, not by an ORM hook).
+    created_by: { type: 'string' },
+    created_on: { type: 'integer' },
+    updated_by: { type: 'string' },
+    updated_on: { type: 'integer' },
+  };
+
+  // All grants for a given grantee (user uid or app name). Used to rebuild the
+  // grantee's policy content so every granted shared path is present/absent.
+  static async listForGrantee(granteeType, granteeId) {
+    return this.list({ where: { granteeType, granteeId } });
+  }
+}
+
+module.exports = { SharedSecretGrant, GRANTEE_TYPES, CAPABILITIES };

nodejs/models/site_join_key.js

@@ -0,0 +1,72 @@
+'use strict';
+
+const crypto = require('crypto');
+const { Model } = require('@simpleworkjs/orm');
+
+// A site join key: the one credential a SPOKE deployment presents to the MASTER
+// to pull a full directory export (LDAP LDIF + resource catalog) when joining
+// (MULTI_SITE_SPEC.md). It works like an agent join key — issued once, shown
+// once, stored hashed, revocable, expirable.
+//
+// The master's POST /api/site/export authenticates callers with this key; the
+// spoke's POST /api/site/join consumes it. The `stj_` prefix distinguishes a
+// site join key from an agent token / `tjk_` agent join key at a glance.
+class SiteJoinKey extends Model {
+	static hashKey(raw) {
+		return crypto.createHash('sha256').update(String(raw || ''), 'utf8').digest('hex');
+	}
+
+	static generateKey() {
+		return 'stj_' + crypto.randomBytes(32).toString('hex');
+	}
+
+	// Resolve a presented key to a usable site join key, or null. Expiry and
+	// revocation are enforced here so no caller can forget one.
+	static async authenticate(rawKey) {
+		if (!rawKey || typeof rawKey !== 'string') return null;
+		const keyHash = this.hashKey(rawKey);
+		const matches = await this.list({ where: { keyHash } });
+		const key = matches && matches[0];
+		if (!key) return null;
+		if (key.revoked) return null;
+		if (key.expires_on && key.expires_on < Math.floor(Date.now() / 1000)) return null;
+		return key;
+	}
+
+	static async issue({ label, createdBy, expiresInDays }) {
+		const raw = this.generateKey();
+		const key = await this.create({
+			id: crypto.randomUUID(),
+			label: label || 'default',
+			keyHash: this.hashKey(raw),
+			keyPrefix: raw.slice(0, 12),
+			revoked: false,
+			created_by: createdBy || null,
+			created_on: Math.floor(Date.now() / 1000),
+			expires_on: expiresInDays ? Math.floor(Date.now() / 1000) + expiresInDays * 86400 : null,
+			use_count: 0
+		});
+		return { key, raw };
+	}
+
+	static fields = {
+		id: { type: 'uuid', primaryKey: true },
+		label: { type: 'string', isRequired: true },
+		keyHash: { type: 'string', isRequired: true },
+		keyPrefix: { type: 'string' },
+		revoked: { type: 'boolean', default: false },
+		created_by: { type: 'string' },
+		created_on: { type: 'integer' },
+		expires_on: { type: 'integer' },
+		use_count: { type: 'integer', default: 0 },
+		last_used_on: { type: 'integer' }
+	};
+
+	toPublic() {
+		const data = this.toJSON ? this.toJSON() : { ...this };
+		delete data.keyHash;
+		return data;
+	}
+}
+
+module.exports = { SiteJoinKey };

nodejs/models/site_spoke.js

@@ -0,0 +1,52 @@
+'use strict';
+
+const crypto = require('crypto');
+const { Model } = require('@simpleworkjs/orm');
+
+// A spoke known to THIS node while it's acting as master — the registry that
+// makes live replication possible. A spoke registers itself here (POST
+// /api/site/spokes, authenticated by the same join key it used to join)
+// right after adopting the master's export, handing over its own reachable
+// endpoint. In return it's issued a `pushToken`: a shared secret the master
+// then presents on every future POST <spoke endpoint>/api/site/resync call.
+//
+// This is a DIFFERENT credential direction than SiteJoinKey: a join key is
+// presented TO the master and only ever needs to be verified (so it's stored
+// hashed, like a password). pushToken is presented BY the master, repeatedly,
+// so it has to be retrievable here -- there is no getting around storing it
+// in plaintext on the master, the same way Webhook.secret is (see
+// services/webhook_emitter.js) for the same reason (an HMAC/bearer credential
+// the sender must keep re-presenting, not a one-time secret only ever
+// verified).
+class SiteSpoke extends Model {
+	static generatePushToken() {
+		return crypto.randomBytes(24).toString('base64url');
+	}
+
+	static fields = {
+		id: { type: 'uuid', primaryKey: true },
+		endpoint: { type: 'string', isRequired: true, unique: true },
+		siteSlug: { type: 'string' },
+		pushToken: { type: 'string', isRequired: true },
+		created_on: { type: 'integer' },
+		last_seen_on: { type: 'integer' },
+		// No-inbound relay (MULTI_SITE_SPEC.md): a spoke with no public IP of
+		// its own reports its WG mesh IP + the public hostname it wants
+		// reached at; the master then best-effort creates a matching relay
+		// route on its own theta-proxy (utils/proxy_client.js). relayNote
+		// records what happened for visibility in the UI -- this automation
+		// is optional/best-effort, never a join requirement.
+		noInbound: { type: 'boolean', default: false },
+		meshIp: { type: 'string' },
+		publicHost: { type: 'string' },
+		relayNote: { type: 'string' }
+	};
+
+	toPublic() {
+		const data = this.toJSON ? this.toJSON() : { ...this };
+		delete data.pushToken;
+		return data;
+	}
+}
+
+module.exports = { SiteSpoke };

nodejs/models/sms.js

@@ -10,6 +10,26 @@ function toE164Digits(number) {
 }
 
 async function send(to, message) {
+	const { PluginInstance } = require('./plugin_instance');
+	const registry = require('../services/plugin_registry');
+	const pluginSecrets = require('../utils/plugin_secrets');
+
+	// @simpleworkjs/orm has no `find` -- the query method is `list({where})`.
+	// `PluginInstance.find(...)` threw "is not a function" on EVERY call into
+	// this sender, so SMS delivery never worked at all: not the test button, not
+	// OTP-by-SMS, not notifications. It failed before it could even fall back to
+	// the direct VoIP.ms path below.
+	const instances = await PluginInstance.list({ where: { category: 'messaging', enabled: true } });
+	if (instances.length > 0) {
+		const inst = instances[0];
+		const manifest = registry.getManifest(inst.pluginType);
+		if (manifest && manifest.sendMessage) {
+			const secrets = await pluginSecrets.read(inst.id).catch(() => ({}));
+			const config = { ...inst.config, ...secrets };
+			return manifest.sendMessage(config, { to, message });
+		}
+	}
+
 	const params = new URLSearchParams({
 		api_username: conf.username,
 		api_password: conf.password,

nodejs/models/token.js

@@ -1,21 +1,17 @@
 'use strict';
 
-const Table = require('.');
-const UUID = function b(a){return a?(a^Math.random()*16>>a/4).toString(16):([1e7]+-1e3+-4e3+-8e3+-1e11).replace(/[018]/g,b)};
+const { Model } = require('@simpleworkjs/orm');
+const crypto = require('crypto');
+const UUID = () => crypto.randomUUID();
 
-
-class Token extends Table{
-	static _key = 'token';
-	static _keyMap = {
-		'created_by': {isRequired: true, type: 'string', min: 3, max: 500},
-		'created_on': {default: function(){return (new Date).getTime()}},
-		'updated_on': {default: function(){return (new Date).getTime()}, always: true},
-		'token': {default: UUID, type: 'string', min: 36, max: 36, isPrivate: true},
-		'is_valid': {default: true, type: 'boolean'},	
-	}
-
-	constructor(...args){
-		super(...args);
+class Token extends Model {
+	static adapterName = 'redis';
+	static fields = {
+		token: { type: 'string', primaryKey: true, default: UUID, isPrivate: true, min: 36, max: 36 },
+		created_by: { isRequired: true, type: 'string', min: 3, max: 500 },
+		created_on: { type: 'integer', default: function(){return (new Date).getTime()} },
+		updated_on: { type: 'integer', default: function(){return (new Date).getTime()}, always: true },
+		is_valid: { default: true, type: 'boolean' }
 	}
 
 	async check(){
@@ -27,12 +23,10 @@ class Token extends Table{
 	}
 }
 
-Token.register();
-
 class AuthToken extends Token{
-	static _keyMap = {
-		...super._keyMap,
-		user: {model: 'User', rel: 'one', localKey: 'created_by'},
+	static fields = {
+		...Token.fields,
+		user: {model: 'User', type: 'hasOne', localKey: 'created_by'},
 	}
 
 	static async create(data){
@@ -41,11 +35,10 @@ class AuthToken extends Token{
 
 	}
 }
-AuthToken.register();
 
 class InviteToken extends Token{
-	static _keyMap = {
-		...super._keyMap,
+	static fields = {
+		...Token.fields,
 		claimed_by:  {default: '__NONE__', isRequired: false, type: 'string'},
 		mail:        {default: '__NONE__', type: 'string'},
 		mail_token:  {default: '__NONE__', type: 'string'},
@@ -67,14 +60,13 @@ class InviteToken extends Token{
 		}
 	}
 }
-InviteToken.register();
 
 class ImpersonationToken extends Token {
-	static _keyMap = {
-		...super._keyMap,
+	static fields = {
+		...Token.fields,
 		target_uid: {isRequired: true, type: 'string', min: 1, max: 200},
 		temp_hash:  {isRequired: true, type: 'string', min: 1, max: 500},
-		expires_at: {default: function(){ return (new Date).getTime() + 7200000 }, type: 'number'},
+		expires_at: {default: function(){ return (new Date).getTime() + 7200000 }, type: 'integer'},
 	}
 
 	get isExpired() {
@@ -86,42 +78,48 @@ class ImpersonationToken extends Token {
 		return this.create(data);
 	}
 }
-ImpersonationToken.register();
 
 class PasswordResetToken extends Token {}
-PasswordResetToken.register();
 
 class OtpToken extends Token {
-	static _keyMap = {
-		...Token._keyMap,
+	static fields = {
+		...Token.fields,
 		uid:        {isRequired: true, type: 'string'},
 		code:       {isRequired: true, type: 'string'},
 		method:     {isRequired: true, type: 'string'},
-		expires_at: {default: function(){ return (new Date).getTime() + 600000 }, type: 'number'},
+		expires_at: {default: function(){ return (new Date).getTime() + 600000 }, type: 'integer'},
 	};
 
 	get isExpired() {
 		return (new Date).getTime() > this.expires_at;
 	}
 
-	// Factory method — named `issue` to avoid shadowing Token's `create(data)`
 	static async issue(uid, method) {
-		const existing = await this.listDetail({uid});
+		const existing = await this.list({where: {uid}});
 		for (const t of existing) {
 			if (t.is_valid) await t.update({is_valid: false});
 		}
-		const code = String(Math.floor(100000 + Math.random() * 900000));
+		const code = String(crypto.randomInt(100000, 1000000));
 		return this.create({uid, code, method, created_by: uid});
 	}
 
 	static async verify(uid, code) {
-		const tokens = await this.listDetail({uid});
+		const tokens = await this.list({where: {uid}});
 		const match = tokens.find(t => t.is_valid && !t.isExpired && t.code === code);
 		if (!match) return null;
 		await match.update({is_valid: false});
 		return match;
 	}
 }
-OtpToken.register();
+class ServiceToken extends Token {
+	static fields = {
+		...Token.fields,
+		resource_id: {isRequired: true, type: 'string'}
+	}
+	
+	static async issue(resource_id, created_by) {
+		return this.create({resource_id, created_by});
+	}
+}
 
-module.exports = {Token, InviteToken, AuthToken, ImpersonationToken, PasswordResetToken, OtpToken};
+module.exports = {Token, InviteToken, AuthToken, ImpersonationToken, PasswordResetToken, OtpToken, ServiceToken};

nodejs/models/user_ldap.js

@@ -9,6 +9,14 @@ const {Token, InviteToken, PasswordResetToken} = require('./token');
 const {Group} = require('./group_ldap');
 const {UserVerification} = require('./verification');
 const conf = require('@simpleworkjs/conf').ldap;
+// Connection + escaping come from the shared @simpleworkjs/ldap package. The
+// wrappers below preserve this file's no-arg call signatures (makeClient() /
+// withClient(fn)) so no call site changes; sso's makeClient passes no
+// tlsOptions, which the shared client forwards as undefined — identical to the
+// previous `new Client({ url: conf.url })`.
+const { makeClient: _makeClient, withClient: _withClient, escapeFilter, escapeDN } = require('@simpleworkjs/ldap');
+const escapeLDAPSearchValue = escapeFilter;
+const escapeLDAPDNValue = escapeDN;
 
 function hashPasswordSSHA512(password) {
 	const salt = crypto.randomBytes(8);
@@ -23,26 +31,11 @@ const cache = new LRUCache({
 });
 
 function makeClient() {
-	return new Client({ url: conf.url });
+	return _makeClient(conf);
 }
 
 async function withClient(fn) {
-	const client = makeClient();
-	try {
-		await client.bind(conf.bindDN, conf.bindPassword);
-		return await fn(client);
-	} finally {
-		await client.unbind().catch(() => {});
-	}
-}
-
-// Helper to escape LDAP filter values (crucial for security)
-function escapeLDAPSearchValue(val) {
-    return val.replace(/\\/g, '\\5c')
-              .replace(/\*/g, '\\2a')
-              .replace(/\(/g, '\\28')
-              .replace(/\)/g, '\\29')
-              .replace(/\0/g, '\\00');
+	return _withClient(conf, fn);
 }
 
 // Compute the next available uid/gidNumber: the highest existing value below
@@ -72,7 +65,8 @@ async function addPosixGroup(client, data){
 
 	data.gidNumber = nextPosixId(groups, 'gidNumber');
 
-	await client.add(`cn=${data.cn},${conf.groupBase}`, {
+	const safeCn = escapeLDAPDNValue(data.cn);
+	await client.add(`cn=${safeCn},${conf.groupBase}`, {
 	  cn: data.cn,
 	  gidNumber: data.gidNumber,
 	  objectclass: [ 'posixGroup', 'top' ]
@@ -94,6 +88,7 @@ async function addPosixAccount(client, data){
 
 	data.uidNumber = nextPosixId(people, 'uidNumber');
 
+	const safeCn = escapeLDAPDNValue(data.cn);
 	const entry = {
 		cn: data.cn,
 		sn: data.sn,
@@ -103,7 +98,6 @@ async function addPosixAccount(client, data){
 		givenName: data.givenName,
 		loginShell: data.loginShell,
 		homeDirectory: data.homeDirectory,
-		userPassword: data.userPassword,
 		description: data.description || ' ',
 		sudoHost: 'ALL',
 		sudoCommand: 'ALL',
@@ -131,7 +125,24 @@ async function addPosixAccount(client, data){
 		entry.dateOfBirth = data.dob;
 	}
 
-	await client.add(`cn=${data.cn},${conf.userBase}`, entry);
+	if (data.location) {
+		entry.l = data.location;
+	}
+
+	// userPassword is optional -- a service account with no password set
+	// simply can't bind (no special enforcement needed, that's the default
+	// LDAP simple-bind behavior for an entry lacking the attribute).
+	if (data.userPassword) {
+		entry.userPassword = data.userPassword;
+	}
+
+	// manager (COSINE, SUP distinguishedName) is naturally multi-valued --
+	// every account gets at least the DN of whoever created it.
+	if (data.manager && [].concat(data.manager).length) {
+		entry.manager = [].concat(data.manager);
+	}
+
+	await client.add(`cn=${safeCn},${conf.userBase}`, entry);
 
 	return data
 
@@ -151,11 +162,14 @@ async function addLdapUser(client, data){
 			data.uid = `${data.givenName[0]}${data.sn}`.toLowerCase();
 		}
 		data.cn = data.uid;
-		data.loginShell = '/bin/bash';
-		data.homeDirectory= `/home/${data.uid}`;
-		data.userPassword = hashPasswordSSHA512(data.userPassword);
+		data.loginShell = data.loginShell || '/bin/bash';
+		data.homeDirectory = data.homeDirectory || `/home/${data.uid}`;
+		if (data.userPassword) {
+			data.userPassword = hashPasswordSSHA512(data.userPassword);
+		} else {
+			delete data.userPassword;
+		}
 
-		console.log('addLdapUser', data)
 		group = await addPosixGroup(client, data);
 		data = await addPosixAccount(client, group);
 
@@ -190,10 +204,19 @@ const user_parse = function(data){
 		data.username = data[conf.userNameAttribute]
 		data.userPassword = undefined;
 	}
+	data.location = data.l ? String(data.l) : '';
 	// Use truthy strings so jq-repeat section blocks ({{#isActive}}) fire correctly
 	data.isActive   = data.pwdAccountLockedTime ? '' : 'active';
 	data.isInactive = data.pwdAccountLockedTime ? 'inactive' : '';
 
+	// manager (COSINE, SUP distinguishedName) and memberOf (from the memberof
+	// overlay) are both multi-valued; ldapts returns a bare string for a
+	// single value and an array for multiple -- normalize both to always be
+	// an array, or app-base.js's `for(let group of user.memberOf)` silently
+	// iterates a single DN string character-by-character instead of once.
+	data.manager   = [].concat(data.manager || []).filter(Boolean);
+	data.memberOf  = [].concat(data.memberOf || []).filter(Boolean);
+
 	return data;
 }
 
@@ -242,6 +265,8 @@ User.listDetail = async function(){
 			serviceAccountDNs = new Set((svcGroup.member || []).map(dn => dn.toLowerCase()));
 		}catch(error){ /* group not seeded yet on an old deployment -- treat as none */ }
 
+		const dnToUid = new Map(searchEntries.map(e => [String(e.dn).toLowerCase(), e.uid]));
+
 		const users = await Promise.all(searchEntries.map(async (entry) => {
 			const rawPassword = entry.userPassword ? entry.userPassword.toString() : '';
 			const isLegacyMD5 = rawPassword.toUpperCase().startsWith('{MD5}');
@@ -269,6 +294,10 @@ User.listDetail = async function(){
 			].filter(Boolean);
 			obj.onboardingRequired = obj.onboardingNeeds.length > 0 ? 'yes' : '';
 			obj.isServiceAccount   = serviceAccountDNs.has(String(obj.dn).toLowerCase()) ? 'yes' : '';
+			obj.managerUids        = obj.manager.map(dn => dnToUid.get(String(dn).toLowerCase()) || dn);
+			// hasSshKey is a boolean flag for the UI -- sshPublicKey may be an array,
+			// and Mustache's {{#sshPublicKey}}...{{/sshPublicKey}} iterates over each item.
+			obj.hasSshKey          = obj.sshPublicKey ? 'yes' : '';
 
 			return obj;
 		}));
@@ -324,6 +353,13 @@ User.get = async function(data, key) {
 
         const verif = await UserVerification.getOrCreate(obj.uid);
 
+        // Same membership check as User.listDetail() -- see the comment there.
+        try{
+            const svcGroup = await Group.get('app_sso_service_account');
+            const serviceAccountDNs = new Set((svcGroup.member || []).map(dn => dn.toLowerCase()));
+            obj.isServiceAccount = serviceAccountDNs.has(String(obj.dn).toLowerCase()) ? 'yes' : '';
+        }catch(error){ obj.isServiceAccount = ''; }
+
         // Auto-flag legacy MD5 password users — persist so subsequent cache hits see it
         if (isLegacyMD5 && !verif.password_must_change) {
             await verif.update({ password_must_change: true });
@@ -421,7 +457,7 @@ User.update = async function(data){
 			}
 		}
 
-		let editableFeilds = ['mobile', 'description'];
+		let editableFeilds = ['mobile', 'description', 'homeDirectory', 'loginShell'];
 
 		await withClient(async (client) => {
 			for(let field of editableFeilds){
@@ -440,6 +476,19 @@ User.update = async function(data){
 			}
 
 			if(data.sshPublicKey){
+				// Ensure the auxiliary objectClass is present before setting the attribute
+				// -- accounts created before ldapPublicKey was added to addPosixAccount's
+				// objectclass list (e.g. the bootstrap admin) won't have it yet.
+				try {
+					await client.modify(this.dn, [
+						new Change({
+							operation: 'add',
+							modification: new Attribute({ type: 'objectClass', values: ['ldapPublicKey'] }),
+						}),
+					]);
+				} catch(e) {
+					if(e.name !== 'TypeOrValueExistsError') throw e;
+				}
 				await client.modify(this.dn, [
 					new Change({
 						operation: 'replace',
@@ -469,6 +518,31 @@ User.update = async function(data){
 				]);
 				this.dateOfBirth = data.dateOfBirth;
 			}
+
+			if(data.location !== undefined){
+				await client.modify(this.dn, [
+					new Change({
+						operation: 'replace',
+						modification: new Attribute({ type: 'l', values: [data.location] }),
+					}),
+				]);
+				this.location = data.location;
+			}
+
+			if(data.manager !== undefined){
+				// Client sends uids; resolve each to a DN before writing --
+				// manager (COSINE, SUP distinguishedName) stores DNs, not uids.
+				const uids = [].concat(data.manager || []).filter(Boolean);
+				const managers = await Promise.all(uids.map(uid => User.get(uid)));
+				const dns = managers.map(u => u.dn);
+				await client.modify(this.dn, [
+					new Change({
+						operation: 'replace',
+						modification: new Attribute({ type: 'manager', values: dns }),
+					}),
+				]);
+				this.manager = dns;
+			}
 		});
 		cache.clear();
 
@@ -537,6 +611,12 @@ User.addByInvite = async function(data){
 
 		data.mail = token.mail;
 
+		// Default manager: whoever sent the invite.
+		try {
+			const inviter = await this.get(token.created_by);
+			data.manager = [inviter.dn];
+		} catch(e) { /* inviter no longer exists -- leave manager unset */ }
+
 		const suggestions = await this.usernameSuggestions(data.givenName, data.sn, data.dob);
 		if (!data.uid || !suggestions.includes(data.uid)) {
 			const err = new Error('Invalid username selection');
@@ -693,7 +773,7 @@ User.setActive = async function(active) {
 				]);
 			} else {
 				await client.modify(this.dn, [
-					new Change({ operation: 'replace', modification: new Attribute({ type: 'pwdAccountLockedTime', values: ['000001010000Z'] }) }),
+					new Change({ operation: 'replace', modification: new Attribute({ type: 'pwdAccountLockedTime', values: ['00000101000000Z'] }) }),
 				]);
 			}
 		});
@@ -708,7 +788,7 @@ User.setActive = async function(active) {
 			throw e;
 		}
 	}
-	this.pwdAccountLockedTime = active ? undefined : '000001010000Z';
+	this.pwdAccountLockedTime = active ? undefined : '00000101000000Z';
 	this.isActive   = active ? 'active' : '';
 	this.isInactive = active ? '' : 'inactive';
 	cache.clear();
@@ -720,6 +800,19 @@ User.addSSHkey = async function(data) {
 	let result;
 	try {
 		await withClient(async (client) => {
+			// Ensure the auxiliary objectClass is present before setting the attribute
+			// -- accounts created before ldapPublicKey was added to addPosixAccount's
+			// objectclass list (e.g. the bootstrap admin) won't have it yet.
+			try {
+				await client.modify(user.dn, [
+					new Change({
+						operation: 'add',
+						modification: new Attribute({ type: 'objectClass', values: ['ldapPublicKey'] }),
+					}),
+				]);
+			} catch(e) {
+				if (e.name !== 'TypeOrValueExistsError') throw e;
+			}
 			await client.modify(user.dn, [
 				new Change({
 					operation: 'add',
@@ -739,6 +832,53 @@ User.addSSHkey = async function(data) {
 	return result;
 };
 
+// Every user gets a personal Unix group of the same name at creation (see
+// addPosixGroup) -- just a GID holder, cn always equal to the user's uid.
+// memberUid (RFC 2307, posixGroup) is a bare username, not a DN, unlike
+// groupOfNames' `member` used by app_sso_* groups in group_ldap.js.
+function personalGroupDN(uid){
+	return `cn=${escapeLDAPDNValue(uid)},${conf.groupBase}`;
+}
+
+User.getPersonalGroupMembers = async function(uid) {
+	try {
+		return await withClient(async (client) => {
+			const res = await client.search(personalGroupDN(uid), {
+				scope: 'base',
+				filter: '(objectClass=posixGroup)',
+				attributes: ['memberUid'],
+			});
+			const entry = res.searchEntries[0];
+			return [].concat((entry && entry.memberUid) || []).filter(Boolean);
+		});
+	} catch(error) {
+		throw error;
+	}
+};
+
+User.addPersonalGroupMember = async function(uid, memberUid) {
+	await this.get(memberUid); // throws UserNotFound if the target uid doesn't exist
+	await withClient(async (client) => {
+		await client.modify(personalGroupDN(uid), [
+			new Change({
+				operation: 'add',
+				modification: new Attribute({ type: 'memberUid', values: [memberUid] }),
+			}),
+		]);
+	});
+};
+
+User.removePersonalGroupMember = async function(uid, memberUid) {
+	await withClient(async (client) => {
+		await client.modify(personalGroupDN(uid), [
+			new Change({
+				operation: 'delete',
+				modification: new Attribute({ type: 'memberUid', values: [memberUid] }),
+			}),
+		]);
+	});
+};
+
 User.invite = async function(data = {}){
 	try{
 		let token = await InviteToken.create({
@@ -759,8 +899,21 @@ User.invite = async function(data = {}){
 
 User.login = async function(data){
 	try{
+		if (!data.uid && !data.username) {
+			let error = new Error('Invalid Credentials, login failed.');
+			error.name = 'LDAPLoginFailed';
+			error.status = 401;
+			throw error;
+		}
 		let user = await this.get(data.uid || data.username);
 
+		if (user.pwdAccountLockedTime) {
+			let error = new Error('Invalid Credentials, login failed.');
+			error.name = 'LDAPLoginFailed';
+			error.status = 401;
+			throw error;
+		}
+
 		const loginClient = makeClient();
 		try {
 			await loginClient.bind(user.dn, data.password);
@@ -771,7 +924,7 @@ User.login = async function(data){
 		return user;
 
 	}catch(error){
-		console.error("USER LOGIN error:", error);
+		console.error("USER LOGIN error:", error.name, error.message);
 		throw error;
 	}
 };

nodejs/models/vault_app_token.js

@@ -0,0 +1,43 @@
+'use strict';
+
+// VaultAppToken — the ACCESSOR of an OpenBao token minted for an external app
+// from the vault UI (Apps tab), so sso can keep the token alive.
+//
+// The token itself is shown ONCE at mint and never stored (a stolen accessor
+// cannot authenticate — it can only look up, renew, or revoke its token, and
+// only the sso broker's policy grants those endpoints). App tokens are minted
+// through the sso-app role as PERIODIC tokens: they live forever, but only if
+// something renews them inside every period window. That something is sso's
+// renewal loop (vault_broker.startAppTokenRenewal), which walks these rows and
+// POSTs auth/token/renew-accessor on a timer — so a downstream app's credential
+// stays valid as long as sso itself is running, with no renewal code needed in
+// the downstream app.
+//
+// One row per app name: re-minting an app's token revokes the previous token
+// via its accessor (no zombie credentials) and replaces the row.
+
+const { Model } = require('@simpleworkjs/orm');
+
+class VaultAppToken extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    // The external app's name — also its policy (app-<name>) and KV namespace
+    // (secret/apps/<name>/). Unique: one live token per app.
+    name: { type: 'string', isRequired: true, unique: true, min: 1, max: 64 },
+    // The minted token's accessor (renew/revoke handle, cannot authenticate).
+    accessor: { type: 'string', isRequired: true, max: 128 },
+    // Renewal bookkeeping, updated by the renewal loop.
+    lastRenewedAt: { type: 'integer' },
+    lastError: { type: 'text' },
+    // Audit stamps (set by the route handler, not by an ORM hook).
+    created_by: { type: 'string' },
+    created_on: { type: 'integer' },
+  };
+
+  static async getByName(name) {
+    const rows = await this.list({ where: { name } });
+    return rows[0] || null;
+  }
+}
+
+module.exports = { VaultAppToken };

nodejs/models/webhook.js

@@ -0,0 +1,15 @@
+const { Model } = require('@simpleworkjs/orm');
+
+class Webhook extends Model {
+  static fields = {
+    id: { type: 'uuid', primaryKey: true },
+    name: { type: 'string', isRequired: true },
+    url: { type: 'string', isRequired: true },
+    events: { type: 'json', default: [] }, // e.g. ['discovery.new_device', 'resource.updated']
+    secret: { type: 'string' },
+    isActive: { type: 'boolean', default: true },
+    created_on: { type: 'integer' },
+  };
+}
+
+module.exports = { Webhook };

nodejs/package.json

@@ -1,7 +1,7 @@
 {
-  "name": "t42-sso-manager",
-  "version": "1.1.1",
-  "private": true,
+  "name": "t42-theta-directory",
+  "version": "2.6.0",
+  "description": "A very simple LDAP management and SSO system",
   "author": [
     {
       "name": "William Mantly",
@@ -11,7 +11,7 @@
   "scripts": {
     "start": "node ./bin/www",
     "dev": "npx nodemon --ignore public/ ./bin/www",
-    "test": "NODE_ENV=test jest --runInBand --forceExit --testTimeout=15000"
+    "test": "NODE_ENV=test jest tests/groups.test.js tests/subtypes.test.js tests/site_join.test.js tests/site_config.test.js tests/site_replicate.test.js tests/proxy_client.test.js tests/reconciler.test.js tests/nmap_plugin.test.js tests/jump_client.test.js --forceExit"
   },
   "jest": {
     "testEnvironment": "node",
@@ -23,26 +23,39 @@
   "dependencies": {
     "@fortawesome/fontawesome-free": "^7.3.0",
     "@popperjs/core": "^2.11.8",
-    "@simpleworkjs/conf": "^1.1.0",
+    "@simpleworkjs/app-stack": "^1.0.0",
+    "@simpleworkjs/bao-conf": "^1.0.1",
+    "@simpleworkjs/conf": "^1.2.0",
+    "@simpleworkjs/directory-schema": "^1.1.0",
+    "@simpleworkjs/frontend": "^0.2.7",
+    "@simpleworkjs/ldap": "^1.0.0",
+    "@simpleworkjs/orm": "^0.2.8",
     "bcrypt": "^6.0.0",
     "bootstrap": "^5.3.8",
+    "bullmq": "^6.0.3",
     "compression": "^1.8.1",
     "ejs": "^3.1.10",
     "express": "^5.2.1",
     "express-rate-limit": "^8.5.2",
     "extend": "^3.0.2",
-    "jq-repeat": "^2.0.1",
-    "jquery": "^3.7.1",
+    "http-proxy-middleware": "^2.0.10",
+    "ioredis": "^6.0.0",
+    "jq-repeat": "^2.2.0",
+    "jquery": "^4.0.0",
     "jsonwebtoken": "^9.0.3",
-    "ldapts": "^8.1.2",
+    "ldapts": "^8.1.8",
     "lru-cache": "^11.5.1",
     "marked": "^9.1.6",
-    "model-redis": "^0.4.0",
+    "model-redis": "^1.6.0",
     "moment": "^2.30.1",
     "mustache": "^4.2.0",
+    "node-fetch": "^2.7.0",
+    "node-nmap": "^4.0.0",
     "nodemailer": "^9.0.0",
     "p2psub": "^0.2.0",
-    "socket.io": "^4.8.3"
+    "socket.io": "^4.8.3",
+    "ws": "^8.21.1",
+    "xss": "^1.0.15"
   },
   "license": "MIT",
   "repository": {

nodejs/patch.diff

@@ -0,0 +1,328 @@
+diff --git a/nodejs/views/directory.ejs b/nodejs/views/directory.ejs
+index c7646a4..411b56f 100644
+--- a/nodejs/views/directory.ejs
++++ b/nodejs/views/directory.ejs
+@@ -3,7 +3,26 @@
+ <div class="container mt-4">
+   <div class="row">
+     <div class="col-12">
+-			<div class="card shadow">
++      <ul class="nav nav-tabs mb-3" id="directoryTabs" role="tablist">
++        <li class="nav-item" role="presentation">
++          <button class="nav-link active" id="directory-tab" data-bs-toggle="tab" data-bs-target="#directory-tab-pane" type="button" role="tab" aria-controls="directory-tab-pane" aria-selected="true">
++            <i class="fa-solid fa-server"></i> Directory
++          </button>
++        </li>
++        <li class="nav-item" role="presentation">
++          <button class="nav-link" id="discovery-tab" data-bs-toggle="tab" data-bs-target="#discovery-tab-pane" type="button" role="tab" aria-controls="discovery-tab-pane" aria-selected="false">
++            <i class="fa-solid fa-network-wired"></i> Discovery
++          </button>
++        </li>
++        <li class="nav-item" role="presentation">
++          <button class="nav-link" id="plugins-tab" data-bs-toggle="tab" data-bs-target="#plugins-tab-pane" type="button" role="tab" aria-controls="plugins-tab-pane" aria-selected="false">
++            <i class="fa-solid fa-plug"></i> Plugins & Scheduler
++          </button>
++        </li>
++      </ul>
++      <div class="tab-content" id="directoryTabsContent">
++        <div class="tab-pane fade show active" id="directory-tab-pane" role="tabpanel" aria-labelledby="directory-tab">
++			<div class="card shadow border-top-0">
+ 				<div class="card-header d-flex flex-wrap justify-content-between align-items-center gap-2">
+ 				    <div>
+     					<i class="fa-solid fa-server"></i> Directory Management
+@@ -74,6 +93,148 @@
+           </table>
+         </div>
+       </div>
++      
++      <!-- Discovery Tab Pane -->
++      <div class="tab-pane fade" id="discovery-tab-pane" role="tabpanel" aria-labelledby="discovery-tab">
++        <div class="card shadow border-top-0">
++          <div class="card-header d-flex flex-wrap justify-content-between align-items-center gap-2">
++            <div>
++              <i class="fa-solid fa-network-wired"></i> Network Discovery Dashboard
++            </div>
++            <div class="d-flex flex-wrap gap-2 align-items-center">
++              <input type="text" id="discovery-search-filter" class="form-control form-control-sm shadow-sm" placeholder="Search resources..." onkeyup="renderDiscoveryTable()" style="width: 250px;">
++              <select id="discovery-filter-managed" class="form-select form-select-sm shadow-sm" onchange="renderDiscoveryTable()" style="width: 150px;">
++                <option value="unmanaged">Unmanaged Only</option>
++                <option value="managed">Managed Only</option>
++                <option value="all">All Resources</option>
++              </select>
++            </div>
++          </div>
++          <div class="card-header actionMessage" style="display:none"></div>
++          <div class="p-3 pb-0 text-muted small border-bottom">
++            <i class="fa-solid fa-circle-info"></i> Auto-discovered network resources. Promote unmanaged devices to track them in the Directory.
++            <a href="/docs/discovery" class="text-reset float-end" title="Help"><i class="fa-solid fa-circle-question"></i></a>
++          </div>
++          <div class="table-responsive">
++            <table class="card-body table table-hover mb-0 align-middle">
++              <thead class="table-light">
++                <tr>
++                  <th class="ps-3">Name / Source</th>
++                  <th>Type</th>
++                  <th>IP Address</th>
++                  <th>Status</th>
++                  <th class="text-end pe-3">Actions</th>
++                </tr>
++              </thead>
++              <tbody id="discovery-list" jq-repeat="discoveryResources">
++                <tr id="discovery-row-{{slug}}">
++                  <td class="ps-3">
++                    <div class="fw-bold">{{name}}</div>
++                    <div class="text-muted small">
++                      <i class="fa-solid fa-plug pe-1"></i> {{#metadata.source}}{{metadata.source}}{{/metadata.source}}{{^metadata.source}}Manual{{/metadata.source}}
++                    </div>
++                  </td>
++                  <td>
++                    <span class="badge bg-secondary">{{kind}}</span>
++                    {{#metadata.subType}}
++                      <span class="badge bg-light text-dark border">{{metadata.subType}}</span>
++                    {{/metadata.subType}}
++                  </td>
++                  <td>
++                    {{#metadata.ip}}<div class="font-monospace small"><i class="fa-solid fa-network-wired pe-1"></i>{{metadata.ip}}</div>{{/metadata.ip}}
++                    {{^metadata.ip}}<span class="text-muted small fst-italic">Unknown IP</span>{{/metadata.ip}}
++                    {{#metadata.interfaces.length}}
++                    <div class="mt-1 small text-muted">
++                      {{#metadata.interfaces}}
++                        <div><i class="fa-solid fa-microchip pe-1"></i> {{mac}} {{#ip}}<span class="text-black-50">({{ip}})</span>{{/ip}}</div>
++                      {{/metadata.interfaces}}
++                    </div>
++                    {{/metadata.interfaces.length}}
++                  </td>
++                  <td>
++                    {{#metadata.managed}}
++                      <span class="badge bg-success rounded-pill px-2"><i class="fa-solid fa-check"></i> Managed</span>
++                    {{/metadata.managed}}
++                    {{^metadata.managed}}
++                      <span class="badge bg-warning text-dark rounded-pill px-2"><i class="fa-solid fa-ghost"></i> Unmanaged</span>
++                    {{/metadata.managed}}
++                  </td>
++                  <td class="text-end pe-3">
++                    {{^metadata.managed}}
++                      <button class="btn btn-sm btn-outline-primary" onclick="promoteResource('{{slug}}')" title="Promote to Managed">
++                        <i class="fa-solid fa-arrow-up-right-dots"></i> Promote
++                      </button>
++                    {{/metadata.managed}}
++                    {{#metadata.managed}}
++                      <button class="btn btn-sm btn-outline-secondary" disabled title="Already Managed">
++                         Promoted
++                      </button>
++                    {{/metadata.managed}}
++                  </td>
++                </tr>
++              </tbody>
++              <tbody id="discovery-empty-state" style="display: none;">
++                <tr>
++                  <td colspan="5" class="text-center py-5 text-muted">
++                    <i class="fa-solid fa-magnifying-glass fs-2 mb-3 text-black-50"></i>
++                    <h5>No resources found</h5>
++                    <p>Check your filters or ensure the discovery agents are running.</p>
++                  </td>
++                </tr>
++              </tbody>
++            </table>
++          </div>
++        </div>
++      </div>
++      
++      <!-- Plugins Tab Pane -->
++      <div class="tab-pane fade" id="plugins-tab-pane" role="tabpanel" aria-labelledby="plugins-tab">
++        <div class="card shadow border-top-0">
++          <div class="card-header d-flex flex-wrap justify-content-between align-items-center gap-2">
++            <div>
++              <i class="fa-solid fa-plug"></i> Plugins & Scheduler
++            </div>
++          </div>
++          <div class="p-3 pb-0 text-muted small border-bottom">
++            <i class="fa-solid fa-circle-info"></i> Manage background tasks and schedules. <a href="/docs/plugins">Learn how to make and use custom plugins</a>.
++          </div>
++          <div class="table-responsive">
++            <table class="card-body table table-hover mb-0 align-middle">
++              <thead class="table-light">
++                <tr>
++                  <th class="ps-3">Plugin Name</th>
++                  <th>Cron Schedule</th>
++                  <th>Status</th>
++                  <th>Actions</th>
++                </tr>
++              </thead>
++              <tbody id="plugins-list" jq-repeat="plugins">
++                <tr>
++                  <td class="ps-3 fw-bold">{{name}}</td>
++                  <td><input type="text" class="form-control form-control-sm font-monospace" id="cron-{{name}}" value="{{cron}}" style="max-width: 150px;"></td>
++                  <td>
++                    {{#enabled}}<span class="badge bg-success">Enabled</span>{{/enabled}}
++                    {{^enabled}}<span class="badge bg-secondary">Disabled</span>{{/enabled}}
++                  </td>
++                  <td>
++                    <button class="btn btn-sm btn-outline-primary" onclick="updatePlugin('{{name}}')" title="Save Schedule">Save</button>
++                    {{#enabled}}<button class="btn btn-sm btn-outline-danger" onclick="togglePlugin('{{name}}', false)">Disable</button>{{/enabled}}
++                    {{^enabled}}<button class="btn btn-sm btn-outline-success" onclick="togglePlugin('{{name}}', true)">Enable</button>{{/enabled}}
++                  </td>
++                </tr>
++              </tbody>
++              <tbody id="plugins-empty-state" style="display: none;">
++                <tr>
++                  <td colspan="4" class="text-center py-4 text-muted">
++                    No plugins configured.
++                  </td>
++                </tr>
++              </tbody>
++            </table>
++          </div>
++        </div>
++      </div>
++      
+     </div>
+   </div>
+ </div>
+@@ -413,14 +574,144 @@
+         const parentEdge = allEdges.find(e => e.childId === r.id);
+         if (parentEdge) {
+            r.parentId = parentEdge.parentId;
+-           const parent = resourcesById[parentEdge.parentId];
++        const parent = resourcesById[parentEdge.parentId];
+            if (parent) r.hostName = parent.name;
+         }
+         rawResources.push(r);
+       }
++       function openAddModal(parent_id, kind) {
++    if(parent_id){
++      $('#newResourceParent').val(parent_id);
++      $('#newResourceKind').val(kind);
++      var currentLabel = "Resource";
++      if(kind === 'Host'){ currentLabel = 'Host'; }
++      else if(kind === 'Site'){ currentLabel = 'Site'; }
++      
++      $('#newResourceLabel').text('Add Child ' + currentLabel);
++    }else{
++      $('#newResourceParent').val('');
++      $('#newResourceKind').val('Host');
++      $('#newResourceLabel').text('Add Resource');
++    }
++    
++    // Clear input
++    $('#newResourceName').val('');
++    $('#addResourceModal').modal('show');
++  }
++
++  // --- DISCOVERY SCRIPTS ---
++  let allDiscoveryResources = [];
++
++  function loadDiscoveryResources() {
++    app.api.get('discovery/resources', function(err, res) {
++      if(err) {
++        $('.actionMessage').html('<div class="alert alert-danger">' + (err.message || 'Error loading resources') + '</div>').show();
++        return;
++      }
++      allDiscoveryResources = res.results || [];
++      renderDiscoveryTable();
++    });
++  }
++
++  function renderDiscoveryTable() {
++    const search = $('#discovery-search-filter').val().toLowerCase();
++    const managedFilter = $('#discovery-filter-managed').val();
++
++    const filtered = allDiscoveryResources.filter(r => {
++      if(search && !r.name.toLowerCase().includes(search) && !r.slug.toLowerCase().includes(search)) return false;
++      const isManaged = !!(r.metadata && r.metadata.managed);
++      if(managedFilter === 'managed' && !isManaged) return false;
++      if(managedFilter === 'unmanaged' && isManaged) return false;
++      return true;
++    });
++
++    $.scope.discoveryResources.empty();
++    for(const r of filtered) {
++      $.scope.discoveryResources.push(r);
++    }
++    
++    if(filtered.length === 0) {
++      $('#discovery-list').hide();
++      $('#discovery-empty-state').show();
++    } else {
++      $('#discovery-list').show();
++      $('#discovery-empty-state').hide();
++    }
++  }
++
++  function promoteResource(slug) {
++    if(!confirm("Are you sure you want to promote this resource? This will generate SSO LDAP groups for it.")) return;
++    app.api.post('discovery/promote/' + slug, {}, function(err, res) {
++      if(err) {
++        alert("Error promoting resource: " + (err.message || err));
++        return;
++      }
++      const resource = allDiscoveryResources.find(r => r.slug === slug);
++      if(resource) {
++        resource.metadata = resource.metadata || {};
++        resource.metadata.managed = true;
++      }
++      $('.actionMessage').html('<div class="alert alert-success alert-dismissible"><button type="button" class="btn-close" data-bs-dismiss="alert"></button>Successfully promoted! Created groups: ' + res.groups.join(', ') + '</div>').show();
++      renderDiscoveryTable();
++      renderTable(); // Also update directory tab
++    });
++  }
++
++  // --- PLUGINS SCRIPTS ---
++  function loadPlugins() {
++    app.api.get('plugins', function(err, res) {
++      if(err) {
++        alert("Error loading plugins: " + (err.message || err));
++        return;
++      }
++      const plugins = res.results || {};
++      const pluginNames = Object.keys(plugins);
+       
+-      renderTable();
++      $.scope.plugins.empty();
++      if(pluginNames.length === 0) {
++        $('#plugins-list').hide();
++        $('#plugins-empty-state').show();
++      } else {
++        pluginNames.forEach(name => {
++          const config = plugins[name];
++          $.scope.plugins.push({
++            name: name,
++            cron: config.cron || '',
++            enabled: config.enabled
++          });
++        });
++        $('#plugins-list').show();
++        $('#plugins-empty-state').hide();
++      }
++    });
++  }
+ 
++  function updatePlugin(name) {
++    const cron = $('#cron-' + name).val();
++    app.api.put('plugins/' + name, {cron: cron}, function(err, res) {
++      if(err) { alert("Failed to save: " + err.message); return; }
++      alert("Saved schedule successfully.");
++    });
++  }
++  
++  function togglePlugin(name, enable) {
++    app.api.put('plugins/' + name, {enabled: enable}, function(err, res) {
++      if(err) { alert("Failed to toggle: " + err.message); return; }
++      loadPlugins();
++    });
++  }
++
++  $(document).ready(function(){
++    renderTable();
++    loadDiscoveryResources();
++    loadPlugins();
++    
++    // Auto-open modal if hash is present
++    if(window.location.hash && window.location.hash.startsWith('#modal-')) {
++      const slug = window.location.hash.replace('#modal-', '');
++      setTimeout(() => openEditModal(slug), 500);
++    }
++  }); 
+       // Type-ahead for the "what can this user reach" lookup. Non-blocking: the
+       // input accepts a free-typed uid whether or not the list ever arrives.
+       loadDirectoryUsers().then(function(users) {

nodejs/plugins/discovery/docker.js

@@ -0,0 +1,131 @@
+const http = require('http');
+
+module.exports = {
+  type: 'docker',
+  category: 'discovery',
+  name: 'Docker Daemon',
+  description: 'Discover running containers and networks from a local or remote Docker daemon.',
+  configSchema: [
+    { key: 'socketPath', label: 'Docker Socket Path', type: 'text', required: false, placeholder: '/var/run/docker.sock' },
+    { key: 'tcpHost', label: 'TCP Host (e.g., http://10.0.0.1:2375)', type: 'url', required: false, placeholder: '' },
+    // Containers in this compose project are the stack's own. They are already
+    // represented in the catalog as services, so they are recorded as managed
+    // and linked to the service they implement instead of arriving as
+    // unmanaged strangers a fresh install has to triage.
+    { key: 'stackProject', label: 'Own compose project', type: 'text', required: false, placeholder: 'theta-suite' },
+    { key: 'hostSlug', label: 'Parent host slug', type: 'text', required: false, placeholder: 'host_<hostname>' },
+    { key: 'location', label: 'Location / Site (optional)', type: 'site_select', required: false, placeholder: 'Default Site' },
+    { key: 'autoPromote', label: 'Auto-promote to Directory', type: 'boolean', required: false, default: false }
+  ],
+
+  validate: async (config) => {
+    if (!config.socketPath && !config.tcpHost) {
+      return { ok: false, error: 'Must provide either socketPath or tcpHost' };
+    }
+    return { ok: true };
+  },
+
+  discover: async (config) => {
+    const isTcp = !!config.tcpHost;
+    
+    const requestOptions = {
+      path: '/containers/json',
+      method: 'GET'
+    };
+
+    if (isTcp) {
+      const url = new URL(config.tcpHost);
+      requestOptions.host = url.hostname;
+      requestOptions.port = url.port || (url.protocol === 'https:' ? 443 : 80);
+      requestOptions.protocol = url.protocol;
+    } else {
+      requestOptions.socketPath = config.socketPath || '/var/run/docker.sock';
+    }
+
+    return new Promise((resolve, reject) => {
+      const req = http.request(requestOptions, (res) => {
+        let body = '';
+        res.on('data', chunk => body += chunk);
+        res.on('end', () => {
+          if (res.statusCode !== 200) {
+            return reject(new Error(`Docker API error: ${res.statusCode} ${body}`));
+          }
+          
+          try {
+            const containers = JSON.parse(body);
+            const resources = [];
+            const edges = [];
+
+            const stackProject = (config.stackProject || '').trim();
+            const hostSlug = (config.hostSlug || '').trim();
+
+            for (const c of containers) {
+              const labels = c.Labels || {};
+              const composeProject = labels['com.docker.compose.project'] || '';
+              const composeService = labels['com.docker.compose.service'] || '';
+              const name = c.Names && c.Names.length > 0 ? c.Names[0].replace(/^\//, '') : c.Id.substring(0, 12);
+
+              // A container id changes every time the container is recreated,
+              // so an id-derived slug made `docker compose up` mint a brand-new
+              // resource on every deploy and orphan the previous one. Prefer
+              // identifiers that survive a recreate: the compose project+service
+              // it belongs to, else its name.
+              const stableKey = composeProject && composeService
+                ? `${composeProject}-${composeService}`
+                : (name || c.Id.substring(0, 12));
+              const slug = `docker-${stableKey.toLowerCase().replace(/[^a-z0-9]+/g, '-').replace(/^-|-$/g, '')}`;
+
+              const ports = (c.Ports || []).map(p => p.PublicPort ? `${p.PublicPort}:${p.PrivatePort}` : `${p.PrivatePort}`).join(', ');
+              const isOwnStack = !!(stackProject && composeProject === stackProject);
+
+              const isIgnored = /openbao|openboa|bao-renewer/i.test(name) || /openbao|openboa|bao-renewer/i.test(composeService);
+
+              if (isIgnored) {
+                continue;
+              }
+
+              resources.push({
+                kind: 'container',
+                name: composeService || name,
+                slug: slug,
+                metadata: {
+                  image: c.Image,
+                  state: c.State,
+                  status: c.Status,
+                  ports: ports,
+                  subType: 'docker',
+                  composeProject: composeProject || undefined,
+                  composeService: composeService || undefined,
+                  containerName: name,
+                  sourceId: stableKey,
+                  ignored: isIgnored ? true : undefined,
+                  // Part of the deployment we are running inside: already
+                  // accounted for, not something to promote.
+                  managed: (isOwnStack || isIgnored) ? true : undefined
+                }
+              });
+
+              // Attach the container to the service it implements when the
+              // catalog already has one under that slug (the bootstrap seeds
+              // `sso-manager`, `proxy`, `jump-host`, … using the same names
+              // compose uses). The reconciler drops an edge whose parent does
+              // not resolve, so an unmatched name is simply not linked.
+              if (isOwnStack && composeService) {
+                edges.push({ parentSlug: composeService, childSlug: slug, relation: 'runs' });
+              } else if (hostSlug) {
+                edges.push({ parentSlug: hostSlug, childSlug: slug, relation: 'hosts' });
+              }
+            }
+
+            resolve({ resources, edges });
+          } catch (e) {
+            reject(new Error(`Failed to parse Docker response: ${e.message}`));
+          }
+        });
+      });
+
+      req.on('error', (e) => reject(new Error(`Docker connection error: ${e.message}`)));
+      req.end();
+    });
+  }
+};

nodejs/plugins/discovery/nmap.js

@@ -0,0 +1,122 @@
+const nmap = require('node-nmap');
+nmap.nmapLocation = "nmap"; // default
+
+module.exports = {
+  // Plugin manifest — see nodejs/services/plugin_registry.js. `targetRange` is
+  // not secret (it's a network range to scan), so it lives in the DB row, not
+  // OpenBao. nmap itself has no credentials to test, so `validate` only checks
+  // the range parses — running a real scan is what `run` does.
+  type: 'nmap',
+  category: 'discovery',
+  name: 'Nmap Network Scan',
+  description: 'Discover hosts and services on a network range using nmap OS + port scans.',
+  configSchema: [
+    { key: 'targetRange', label: 'Target Range', type: 'text', required: true, placeholder: '192.168.1.0/24' },
+    { key: 'location', label: 'Location / Site (optional)', type: 'site_select', required: false, placeholder: 'Default Site' },
+    { key: 'autoPromote', label: 'Auto-promote to Directory', type: 'boolean', required: false, default: false }
+  ],
+
+  validate: async (config) => {
+    const { targetRange } = config;
+    if (!targetRange) return { ok: false, error: 'Missing targetRange' };
+    // nmap accepts CIDR (a.b.c.d/24), ranges (a.b.c.d-50), and host lists. We
+    // only sanity-check shape here — reject anything with shell metacharacters
+    // or whitespace, since node-nmap passes this straight to the nmap binary.
+    if (/\s|[;|&$`<>]/.test(targetRange)) {
+      return { ok: false, error: 'targetRange must not contain whitespace or shell metacharacters' };
+    }
+    return { ok: true };
+  },
+
+  discover: async (config) => {
+    const { targetRange } = config;
+    if (!targetRange) throw new Error("Missing targetRange for Nmap");
+
+    return new Promise((resolve, reject) => {
+      // OsAndPortScan requires root (for -O). NmapScan does a basic port scan (TCP connect if non-root).
+      // Pass custom arguments in constructor so node-nmap includes them before spawning nmap process.
+      // -Pn: treat all hosts as online (skip ping/ARP host discovery which fails inside Docker containers NAT/bridge)
+      // -sT: TCP connect scan (unprivileged scan compatible with container environments)
+      // -F: fast scan (100 top ports)
+      // --min-rate 100: speed up scan rate
+      const customFlags = ['-Pn', '-sT', '-F', '--min-rate', '100'];
+      const scan = new nmap.NmapScan(targetRange, customFlags);
+      
+      if (config.log) config.log(`Starting nmap scan: ${scan.command.join(' ')}`);
+
+      scan.on('complete', function(data) {
+        if (config.log) config.log(`Scan complete. Found ${data ? data.length : 0} hosts.`);
+        const resources = [];
+        const edges = [];
+
+        for (const host of data) {
+          if (!host.ip) continue;
+          const hostId = host.mac ? host.mac.replace(/:/g, '') : host.ip.replace(/\\./g, '_');
+          const hostSlug = `nmap-host-${hostId}`;
+          
+          const interfaces = [{ mac: host.mac || null, ip: host.ip }];
+          
+          resources.push({
+            kind: 'host',
+            name: host.hostname || host.ip,
+            slug: hostSlug,
+            metadata: { interfaces, os: host.osNmap }
+          });
+
+          if (host.openPorts && host.openPorts.length > 0) {
+            for (const port of host.openPorts) {
+              const svcSlug = `nmap-svc-${hostId}-${port.port}`;
+              resources.push({
+                kind: 'service',
+                name: `${port.service} on ${port.port}`,
+                slug: svcSlug,
+                metadata: { port: port.port, protocol: port.protocol }
+              });
+              edges.push({ parentSlug: hostSlug, childSlug: svcSlug, relation: 'exposes' });
+            }
+          }
+        }
+        resolve({ resources, edges });
+      });
+
+      scan.on('error', function(error) {
+        // node-nmap's spawn-missing-binary message ("NMAP not found at command
+        // location: nmap") is opaque to an admin reading lastError. Translate
+        // it into something actionable. (The Dockerfile installs nmap in the
+        // app image; this only fires if someone runs outside the container or
+        // strips the package.)
+        var msg = (error && error.message) || String(error);
+        if (/nmap.*not found|command location/i.test(msg)) {
+          reject(new Error('nmap binary not installed in the container image (rebuild with Dockerfile.openldap, which apk-adds nmap)'));
+          return;
+        }
+        // node-nmap (node_modules/node-nmap/index.js) treats ANY stderr
+        // output from the nmap binary as a fatal scan error -- including
+        // nmap's own benign RTT timing-calibration warnings ("RTTVAR has
+        // grown to over N seconds, decreasing to M"), which it prints
+        // *during* a scan that goes on to complete normally. That means a
+        // scan that actually succeeded (valid XML already sitting in
+        // scan.rawData) got thrown away and reported as a failed run with
+        // zero hosts discovered -- not just a noisy log line. Recover by
+        // manually re-running node-nmap's own XML-parse-then-complete path
+        // (rawDataHandler -> scanComplete -> the 'complete' listener above)
+        // when the "error" is this specific known-benign nmap message and
+        // there's actually output to parse. A genuine XML parse failure
+        // re-emits 'error' with a different message, which falls through to
+        // reject() below same as before -- this only widens the recovery
+        // path, it doesn't swallow real failures.
+        if (/RTTVAR has grown/i.test(msg) && scan.rawData) {
+          scan.rawDataHandler(scan.rawData);
+          return;
+        }
+        reject(error);
+      });
+
+      scan.startScan();
+    });
+  },
+
+  // Generalized plugin contract alias for `discover`. See proxmox.js for why
+  // this references module.exports rather than `this`.
+  run: async (config) => module.exports.discover(config)
+};

nodejs/plugins/discovery/proxmox.js

@@ -0,0 +1,400 @@
+const fetch = require('node-fetch');
+const https = require('https');
+
+// Custom agent to bypass self-signed certs typical in Proxmox
+const agent = new https.Agent({
+  rejectUnauthorized: false
+});
+
+// Accumulates a guest's NICs, keyed by MAC, merging what several Proxmox
+// endpoints each know a piece of: the guest agent knows MAC+IP together, the
+// VM/LXC config knows the MAC even while the guest is stopped, and the LXC
+// interfaces endpoint knows the DHCP-assigned IP. Keying by MAC is what keeps
+// the pairing honest -- the previous code collected MACs and IPs into two flat
+// lists and zipped them by index, which mismatched them on any multi-NIC guest.
+class Interfaces {
+  constructor() { this.byMac = new Map(); this.anonymous = []; }
+
+  // Interfaces that belong to something running INSIDE the guest -- container
+  // engines, overlay networks, VPNs -- rather than to the guest itself. A
+  // Home Assistant VM reported 16 of these (docker0, hassio, 14x veth*)
+  // alongside its one real NIC, which is noise in the directory and, worse,
+  // gives the reconciler a pile of 172.x addresses to match unrelated hosts on.
+  // Only applied to guests; a hypervisor's own bridges are how you reach it.
+  static VIRTUAL_IFACE_RE = /^(lo|docker\d*|hassio|veth|br-|virbr|tap|fwbr|fwln|fwpr|cni|flannel|cali|kube|weave|zt|tailscale|wg|tun|utun)/i;
+
+  static isVirtualName(name) {
+    return !!name && Interfaces.VIRTUAL_IFACE_RE.test(name);
+  }
+
+  // A udev "predictable" name of the form enx<12 hex> encodes the MAC. It is
+  // the only place the Proxmox node network API exposes a physical NIC's MAC
+  // (/nodes/{node}/network carries no hwaddr field at all), so parse it out
+  // rather than leaving every hypervisor MAC-less.
+  static macFromIfaceName(name) {
+    const m = /^enx([0-9a-f]{12})$/i.exec(name || '');
+    if (!m) return null;
+    return m[1].toLowerCase().match(/.{2}/g).join(':');
+  }
+
+  static normalizeMac(mac) {
+    const m = (mac || '').toLowerCase().trim();
+    if (!/^([0-9a-f]{2}:){5}[0-9a-f]{2}$/.test(m)) return null;
+    if (m === '00:00:00:00:00:00') return null;
+    return m;
+  }
+
+  // `ips` are the addresses observed on this one NIC (may be empty for a
+  // stopped guest, where only the MAC is known).
+  add(mac, ips, name) {
+    const key = Interfaces.normalizeMac(mac);
+    const addrs = (ips || []).filter(Boolean);
+    if (!key) {
+      // An IP with no usable MAC is still worth keeping; a NIC with neither is not.
+      if (addrs.length) this.anonymous.push({ mac: null, ip: addrs[0], ips: addrs, name: name || null });
+      return;
+    }
+    const existing = this.byMac.get(key);
+    if (existing) {
+      for (const ip of addrs) if (!existing.ips.includes(ip)) existing.ips.push(ip);
+      existing.ip = existing.ips[0] || null;
+      if (!existing.name && name) existing.name = name;
+      return;
+    }
+    this.byMac.set(key, { mac: key, ip: addrs[0] || null, ips: addrs, name: name || null });
+  }
+
+  toArray() { return [...this.byMac.values(), ...this.anonymous]; }
+
+  // The address/MAC the directory shows in its single-value columns, and what
+  // the reconciler matches on. Prefer a NIC that actually has an address.
+  primaryIp() {
+    const withIp = this.toArray().find(i => i.ip);
+    return withIp ? withIp.ip : null;
+  }
+
+  primaryMac() {
+    const withIp = this.toArray().find(i => i.ip && i.mac);
+    if (withIp) return withIp.mac;
+    const first = this.toArray().find(i => i.mac);
+    return first ? first.mac : null;
+  }
+}
+
+module.exports = {
+  // Plugin manifest — see nodejs/services/plugin_registry.js. `configSchema`
+  // drives the admin UI form and validation; fields flagged `secret:true` are
+  // stored in OpenBao (secret/plugins/<instance-id>/conf), never in the DB.
+  type: 'proxmox',
+  category: 'discovery',
+  name: 'Proxmox VE',
+  description: 'Discover VMs, containers, and hypervisor nodes from a Proxmox VE API endpoint.',
+  configSchema: [
+    { key: 'url',         label: 'API URL',      type: 'url',      required: true, placeholder: 'https://pve.example:8006' },
+    { key: 'tokenId',     label: 'Token ID',     type: 'text',     required: true, placeholder: 'user@pam!token' },
+    { key: 'tokenSecret', label: 'Token Secret', type: 'password', required: true, secret: true },
+    { key: 'location',    label: 'Location / Site (optional)', type: 'site_select', required: false, placeholder: 'Default Site' },
+    { key: 'autoPromote', label: 'Auto-promote to Directory', type: 'boolean', required: false, default: false }
+  ],
+
+  // "Test" button in the UI: hit the unauthenticated version endpoint with the
+  // API token to confirm the URL + token are valid before scheduling runs.
+  validate: async (config) => {
+    const { url, tokenId, tokenSecret } = config;
+    if (!url || !tokenId || !tokenSecret) return { ok: false, error: 'Missing url, tokenId, or tokenSecret' };
+    try {
+      const res = await fetch(`${url}/api2/json/version`, { headers: { 'Authorization': `PVEAPIToken=${tokenId}=${tokenSecret}` }, agent });
+      if (!res.ok) return { ok: false, error: `Proxmox API rejected the token (${res.status})` };
+      return { ok: true };
+    } catch (err) {
+      return { ok: false, error: err.message };
+    }
+  },
+
+  discover: async (config) => {
+    let { url, tokenId, tokenSecret } = config;
+    if (!url || !tokenId || !tokenSecret) {
+      throw new Error("Missing Proxmox config");
+    }
+
+    const headers = {
+      'Authorization': `PVEAPIToken=${tokenId}=${tokenSecret}`
+    };
+    
+    // Ensure URL has no trailing slash
+    url = url.endsWith('/') ? url.slice(0, -1) : url;
+
+    const resources = [];
+    const edges = [];
+
+    // 0. The Proxmox endpoint itself. Without it a multi-node cluster produces
+    // several unrelated roots in the Directory tree and nothing says where any
+    // of them came from. Every node discovered below is parented to this, so
+    // one endpoint == one subtree.
+    const endpointHost = (() => {
+      try { return new URL(url).hostname; } catch (e) { return url.replace(/^https?:\/\//, '').split('/')[0]; }
+    })();
+    const clusterName = await (async () => {
+      // /cluster/status names the cluster when one exists; a standalone node
+      // has no cluster entry, in which case the endpoint hostname is the name.
+      try {
+        const res = await fetch(`${url}/api2/json/cluster/status`, { headers, agent });
+        if (!res.ok) return null;
+        const entry = ((await res.json()).data || []).find(d => d.type === 'cluster');
+        return entry ? entry.name : null;
+      } catch (e) { return null; }
+    })();
+
+    const endpointSlug = `pve-${endpointHost.toLowerCase().replace(/[^a-z0-9]+/g, '-').replace(/^-|-$/g, '')}`;
+    resources.push({
+      kind: 'host',
+      name: clusterName || `Proxmox (${endpointHost})`,
+      slug: endpointSlug,
+      metadata: {
+        subType: 'proxmox',
+        address: url,
+        os: 'Proxmox VE',
+        isProduction: true,
+        sourceId: url,
+        // Deliberately NO `ip`/`interfaces`: this resource stands for the
+        // cluster (the API endpoint), not for a machine. Giving it the address
+        // it is reached at made the reconciler match it to the very node that
+        // answers on that address -- the endpoint and the node collapsed into
+        // one row, which then became its own parent. The cluster is identified
+        // by slug + sourceId instead, which nothing else can collide with.
+        interfaces: []
+      }
+    });
+
+    // 1. Get Nodes
+    const resNodes = await fetch(`${url}/api2/json/nodes`, { headers, agent });
+    if(!resNodes.ok) {
+        const errText = await resNodes.text();
+        throw new Error(`Proxmox API error on nodes: ${resNodes.status} ${errText}`);
+    }
+    const nodes = (await resNodes.json()).data;
+
+    for (const node of nodes) {
+      // An offline node is still a real hypervisor that belongs in the
+      // directory -- skipping it entirely used to make it look decommissioned
+      // and let the reconciler's garbage collector archive it after a week of
+      // downtime. Record it, mark it down, and skip only the guest enumeration
+      // (which needs the node to answer).
+      const online = node.status === 'online';
+
+      const nodeSlug = `pve-node-${node.node}`;
+
+      // A hypervisor with no address is not actionable. Read its bridges/NICs
+      // so the node lands in the directory reachable and MAC-identified like
+      // any other host. Unlike a guest, a node's bridges are kept: vmbrN is
+      // normally the address you actually reach the hypervisor on.
+      const nodeIfaces = new Interfaces();
+      try {
+        const netRes = online
+          ? await fetch(`${url}/api2/json/nodes/${node.node}/network`, { headers, agent })
+          : { ok: false };
+        if (netRes.ok) {
+          const ifaceList = (await netRes.json()).data || [];
+          for (const iface of ifaceList) {
+            if (iface.iface === 'lo') continue;
+            const ip = iface.address || iface.cidr;
+            // This endpoint has no hwaddr field, so the MAC has to be recovered
+            // from a predictable interface name -- either this interface's own
+            // or, for a bridge, one of the physical ports beneath it.
+            let mac = Interfaces.macFromIfaceName(iface.iface);
+            if (!mac) {
+              for (const alt of (iface.altnames || [])) {
+                mac = Interfaces.macFromIfaceName(alt);
+                if (mac) break;
+              }
+            }
+            if (!mac && iface.bridge_ports) {
+              for (const port of String(iface.bridge_ports).split(/\s+/).filter(Boolean)) {
+                mac = Interfaces.macFromIfaceName(port);
+                if (mac) break;
+                // The port may itself only carry the MAC in an altname.
+                const portDef = ifaceList.find(i => i.iface === port);
+                for (const alt of ((portDef && portDef.altnames) || [])) {
+                  mac = Interfaces.macFromIfaceName(alt);
+                  if (mac) break;
+                }
+                if (mac) break;
+              }
+            }
+            nodeIfaces.add(mac || iface.hwaddr, ip ? [String(ip).split('/')[0]] : [], iface.iface);
+          }
+        }
+      } catch (e) {}
+
+      resources.push({
+        kind: 'host',
+        name: node.node,
+        slug: nodeSlug,
+        metadata: {
+          subType: 'hypervisor',
+          os: 'Proxmox VE',
+          isProduction: true,
+          status: node.status,
+          sourceId: `${node.node}`,
+          node: node.node,
+          interfaces: nodeIfaces.toArray(),
+          macAddress: nodeIfaces.primaryMac(),
+          ip: nodeIfaces.primaryIp()
+        }
+      });
+      edges.push({ parentSlug: endpointSlug, childSlug: nodeSlug, relation: 'hosts' });
+
+      // Everything below asks the node itself; an offline node answers none of
+      // it, and its guests are already recorded from previous runs.
+      if (!online) continue;
+
+      // 2. Get VMs for this node
+      const resVms = await fetch(`${url}/api2/json/nodes/${node.node}/qemu`, { headers, agent });
+      const vms = resVms.ok ? ((await resVms.json()).data || []) : [];
+
+      for (const vm of vms) {
+        const vmSlug = `vm-${vm.vmid}`;
+        const isTemplate = vm.template === 1;
+        
+        const ifaces = new Interfaces();
+
+        // Enrich from QEMU guest agent if running. The agent is the only source
+        // that knows which IP sits on which NIC, so pair them here rather than
+        // accumulating two flat lists (zipping those by index attributed IPs to
+        // the wrong MAC on any guest with more than one NIC).
+        if (vm.status === 'running') {
+            try {
+                const agentRes = await fetch(`${url}/api2/json/nodes/${node.node}/qemu/${vm.vmid}/agent/network-get-interfaces`, { headers, agent });
+                if (agentRes.ok) {
+                    const agentData = (await agentRes.json()).data;
+                    if (agentData && agentData.result) {
+                        for (const iface of agentData.result) {
+                            // Docker bridges, veth pairs and VPN tunnels are the
+                            // guest's own plumbing, not NICs of the guest.
+                            if (Interfaces.isVirtualName(iface.name)) continue;
+                            const ips = (iface['ip-addresses'] || [])
+                                .filter(ip => ip['ip-address-type'] === 'ipv4' && ip['ip-address'] !== '127.0.0.1')
+                                .map(ip => ip['ip-address']);
+                            ifaces.add(iface['hardware-address'], ips, iface.name);
+                        }
+                    }
+                }
+            } catch(e) {}
+        }
+
+        // Enrich from VM config: the MAC is declared there whether or not the
+        // guest agent answered, so a stopped VM still gets a stable identity.
+        try {
+            const configRes = await fetch(`${url}/api2/json/nodes/${node.node}/qemu/${vm.vmid}/config`, { headers, agent });
+            if (configRes.ok) {
+                const confData = (await configRes.json()).data;
+                for (let i = 0; i < 10; i++) {
+                    if (confData[`net${i}`]) {
+                        const m = confData[`net${i}`].match(/(?:virtio|e1000e?|rtl8139|vmxnet3)=([0-9a-fA-F:]{17})/);
+                        if(m) ifaces.add(m[1], [], `net${i}`);
+                    }
+                }
+            }
+        } catch(e) {}
+
+        const interfaces = ifaces.toArray();
+
+        resources.push({
+          kind: isTemplate ? 'template' : 'host',
+          name: vm.name || `VM ${vm.vmid}`,
+          slug: vmSlug,
+          metadata: {
+            subType: isTemplate ? 'template' : 'vm',
+            vmid: vm.vmid,
+            // The Proxmox-side identity, so a resource can be traced back to the
+            // exact guest on the exact node it was discovered from.
+            sourceId: `${node.node}/qemu/${vm.vmid}`,
+            node: node.node,
+            isProduction: vm.status === 'running',
+            interfaces,
+            macAddress: ifaces.primaryMac(),
+            ip: ifaces.primaryIp()
+          }
+        });
+        edges.push({ parentSlug: nodeSlug, childSlug: vmSlug, relation: 'hosts' });
+      }
+
+      // 3. Get LXCs for this node
+      const resLxcs = await fetch(`${url}/api2/json/nodes/${node.node}/lxc`, { headers, agent });
+      const lxcs = resLxcs.ok ? ((await resLxcs.json()).data || []) : [];
+
+      for (const lxc of lxcs) {
+        const lxcSlug = `lxc-${lxc.vmid}`;
+        const isTemplate = lxc.template === 1;
+        
+        const ifaces = new Interfaces();
+
+        // Enrich from LXC config. Each netN line carries its own hwaddr and ip,
+        // so read them off the same line instead of into parallel lists.
+        try {
+            const configRes = await fetch(`${url}/api2/json/nodes/${node.node}/lxc/${lxc.vmid}/config`, { headers, agent });
+            if (configRes.ok) {
+                const confData = (await configRes.json()).data;
+                for (let i = 0; i < 10; i++) {
+                    const line = confData[`net${i}`];
+                    if (!line) continue;
+                    const hwMatch = line.match(/hwaddr=([0-9a-fA-F:]{17})/);
+                    // `ip=` is either a CIDR address or the literal `dhcp`/`manual`.
+                    const ipMatch = line.match(/\bip=(\d+\.\d+\.\d+\.\d+)/);
+                    const nameMatch = line.match(/\bname=([^,]+)/);
+                    if (hwMatch || ipMatch) {
+                        ifaces.add(hwMatch && hwMatch[1], ipMatch ? [ipMatch[1]] : [], nameMatch ? nameMatch[1] : `net${i}`);
+                    }
+                }
+            }
+        } catch(e) {}
+
+        // A DHCP-configured container has no IP in its config. Ask the running
+        // container's interface list so it lands in the directory addressable
+        // instead of as an IP-less row.
+        if (lxc.status === 'running' && !ifaces.primaryIp()) {
+            try {
+                const ifRes = await fetch(`${url}/api2/json/nodes/${node.node}/lxc/${lxc.vmid}/interfaces`, { headers, agent });
+                if (ifRes.ok) {
+                    for (const iface of ((await ifRes.json()).data || [])) {
+                        if (Interfaces.isVirtualName(iface.name)) continue;
+                        const ip = (iface.inet || '').split('/')[0];
+                        ifaces.add(iface.hwaddr, ip ? [ip] : [], iface.name);
+                    }
+                }
+            } catch(e) {}
+        }
+
+        const interfaces = ifaces.toArray();
+
+        resources.push({
+          kind: isTemplate ? 'template' : 'host',
+          name: lxc.name || `LXC ${lxc.vmid}`,
+          slug: lxcSlug,
+          metadata: {
+            subType: isTemplate ? 'template' : 'lxc',
+            vmid: lxc.vmid,
+            sourceId: `${node.node}/lxc/${lxc.vmid}`,
+            node: node.node,
+            isProduction: lxc.status === 'running',
+            interfaces,
+            macAddress: ifaces.primaryMac(),
+            ip: ifaces.primaryIp()
+          }
+        });
+        edges.push({ parentSlug: nodeSlug, childSlug: lxcSlug, relation: 'hosts' });
+      }
+    }
+
+    return { resources, edges };
+  },
+
+  // The generalized plugin contract calls `run`; the discovery plugins keep
+  // `discover` as their implementation name for back-compat, and `run` is just
+  // an alias. Referenced via module.exports (not `this`) so it survives being
+  // detached and called as a bare function reference.
+  run: async (config) => module.exports.discover(config),
+
+  // Exported for unit tests only -- not part of the plugin contract.
+  _Interfaces: Interfaces
+};

nodejs/plugins/discovery/unifi.js

@@ -0,0 +1,136 @@
+const fetch = require('node-fetch');
+const https = require('https');
+
+const agent = new https.Agent({
+  rejectUnauthorized: false
+});
+
+module.exports = {
+  // Plugin manifest — see nodejs/services/plugin_registry.js. `password` is
+  // secret and stored in OpenBao (secret/plugins/<instance-id>/conf).
+  type: 'unifi',
+  category: 'discovery',
+  name: 'UniFi Network',
+  description: 'Discover UniFi network devices and clients from a UniFi Controller / UDM endpoint.',
+  configSchema: [
+    { key: 'url',      label: 'Controller URL', type: 'url',      required: true, placeholder: 'https://unifi.example:8443' },
+    { key: 'user',     label: 'Username',       type: 'text',     required: true },
+    { key: 'password', label: 'Password',       type: 'password', required: true, secret: true },
+    { key: 'location', label: 'Location / Site (optional)', type: 'site_select', required: false, placeholder: 'Default Site' },
+    { key: 'autoPromote', label: 'Auto-promote to Directory', type: 'boolean', required: false, default: false }
+  ],
+
+  // "Test": attempt the UDM login (falls back to the legacy controller login);
+  // succeeds only if one of the two login endpoints returns 200.
+  validate: async (config) => {
+    const { url, user, password } = config;
+    if (!url || !user || !password) return { ok: false, error: 'Missing url, user, or password' };
+    try {
+      let loginRes = await fetch(`${url}/api/auth/login`, {
+        method: 'POST', headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({ username: user, password }), agent
+      });
+      if (!loginRes.ok) {
+        loginRes = await fetch(`${url}/api/login`, {
+          method: 'POST', headers: { 'Content-Type': 'application/json' },
+          body: JSON.stringify({ username: user, password }), agent
+        });
+      }
+      if (!loginRes.ok) return { ok: false, error: `UniFi auth failed (${loginRes.status})` };
+      return { ok: true };
+    } catch (err) {
+      return { ok: false, error: err.message };
+    }
+  },
+
+  discover: async (config) => {
+    const { url, user, password } = config;
+    if (!url || !user || !password) {
+      throw new Error("Missing Unifi config");
+    }
+
+    // 1. Authenticate
+    let loginRes = await fetch(`${url}/api/auth/login`, {
+      method: 'POST',
+      headers: { 'Content-Type': 'application/json' },
+      body: JSON.stringify({ username: user, password }),
+      agent
+    });
+
+    let isUdm = true;
+    if (!loginRes.ok) {
+      loginRes = await fetch(`${url}/api/login`, {
+        method: 'POST',
+        headers: { 'Content-Type': 'application/json' },
+        body: JSON.stringify({ username: user, password }),
+        agent
+      });
+      isUdm = false;
+    }
+
+    if (!loginRes.ok) {
+      throw new Error(`Unifi auth failed: ${loginRes.status}`);
+    }
+
+    const cookie = loginRes.headers.get('set-cookie');
+    // UniFi often requires the CSRF token from the cookie
+    let csrf = '';
+    if (cookie) {
+      const match = cookie.match(/csrf_token=([^;]+)/);
+      if (match) csrf = match[1];
+    }
+    const headers = { 'Cookie': cookie, 'X-Csrf-Token': csrf };
+
+    const resources = [];
+    const edges = [];
+
+    const basePath = isUdm ? '/proxy/network' : '';
+
+    // 2. Get Devices (Switches/APs)
+    const devRes = await fetch(`${url}${basePath}/api/s/default/stat/device`, { headers, agent });
+    const devData = (await devRes.json()).data || [];
+
+    for (const dev of devData) {
+      const devSlug = `unifi-device-${dev.mac.replace(/:/g, '')}`;
+      resources.push({
+        kind: 'network_device',
+        name: dev.name || dev.model,
+        slug: devSlug,
+        metadata: {
+          make: 'Ubiquiti',
+          model: dev.model,
+          firmware: dev.version,
+          interfaces: [{ mac: dev.mac, ip: dev.ip }]
+        }
+      });
+    }
+
+    // 3. Get Clients
+    const clientRes = await fetch(`${url}${basePath}/api/s/default/stat/sta`, { headers, agent });
+    const clientData = (await clientRes.json()).data || [];
+
+    for (const client of clientData) {
+      const clientSlug = `unifi-client-${client.mac.replace(/:/g, '')}`;
+      resources.push({
+        kind: 'host', // Or unmanaged_device initially
+        name: client.hostname || client.name || client.mac,
+        slug: clientSlug,
+        metadata: {
+          interfaces: [{ mac: client.mac, ip: client.ip }]
+        }
+      });
+      
+      // If we know which switch/AP it's on
+      if (client.ap_mac) {
+        const apSlug = `unifi-device-${client.ap_mac.replace(/:/g, '')}`;
+        edges.push({ parentSlug: apSlug, childSlug: clientSlug, relation: 'connected_to' });
+      }
+    }
+
+    return { resources, edges };
+  },
+
+  // Generalized plugin contract alias for `discover`. See proxmox.js for why
+  // this references module.exports rather than `this`.
+  run: async (config) => module.exports.discover(config)
+};

nodejs/plugins/messaging/twilio.js

@@ -0,0 +1,61 @@
+const https = require('https');
+
+module.exports = {
+  type: 'twilio',
+  category: 'messaging',
+  name: 'Twilio SMS',
+  description: 'Send SMS messages (like 2FA codes) via Twilio.',
+  
+  configSchema: [
+    { key: 'accountSid', label: 'Account SID', type: 'text', required: true },
+    { key: 'authToken', label: 'Auth Token', type: 'password', required: true, secret: true },
+    { key: 'fromNumber', label: 'From Phone Number', type: 'text', required: true, placeholder: '+15551234567' }
+  ],
+
+  validate: async (config) => {
+    if (!config.accountSid || !config.authToken) return { ok: false, error: 'Missing credentials' };
+    if (!config.fromNumber) return { ok: false, error: 'Missing fromNumber' };
+    return { ok: true };
+  },
+
+  sendMessage: async (config, payload) => {
+    const { to, message } = payload;
+    if (!to || !message) throw new Error("Missing 'to' or 'message' in payload");
+
+    const data = new URLSearchParams();
+    data.append('To', to);
+    data.append('From', config.fromNumber);
+    data.append('Body', message);
+
+    const postData = data.toString();
+
+    const options = {
+      hostname: 'api.twilio.com',
+      port: 443,
+      path: `/2010-04-01/Accounts/${config.accountSid}/Messages.json`,
+      method: 'POST',
+      headers: {
+        'Authorization': 'Basic ' + Buffer.from(config.accountSid + ':' + config.authToken).toString('base64'),
+        'Content-Type': 'application/x-www-form-urlencoded',
+        'Content-Length': Buffer.byteLength(postData)
+      }
+    };
+
+    return new Promise((resolve, reject) => {
+      const req = https.request(options, (res) => {
+        let body = '';
+        res.on('data', chunk => body += chunk);
+        res.on('end', () => {
+          if (res.statusCode >= 200 && res.statusCode < 300) {
+            resolve(JSON.parse(body));
+          } else {
+            reject(new Error(`Twilio API Error: ${res.statusCode} ${body}`));
+          }
+        });
+      });
+      req.on('error', reject);
+      req.write(postData);
+      req.end();
+    });
+  }
+};

nodejs/plugins/messaging/webhook.js

@@ -0,0 +1,79 @@
+const https = require('https');
+const http = require('http');
+
+module.exports = {
+  type: 'webhook',
+  category: 'messaging',
+  name: 'Universal REST Webhook',
+  description: 'Send a generic HTTP POST request with a custom JSON payload. Variables {{to}} and {{message}} will be replaced.',
+  
+  configSchema: [
+    { key: 'url', label: 'Webhook URL', type: 'url', required: true, placeholder: 'https://api.example.com/send' },
+    { key: 'method', label: 'HTTP Method', type: 'text', required: true, placeholder: 'POST' },
+    { key: 'headers', label: 'Custom Headers (JSON)', type: 'text', required: false, placeholder: '{"Authorization": "Bearer ...", "Content-Type": "application/json"}' },
+    { key: 'payloadTemplate', label: 'Payload Template', type: 'text', required: true, placeholder: '{"recipient": "{{to}}", "text": "{{message}}"}' },
+    { key: 'apiSecret', label: 'API Secret / Auth Token', type: 'password', required: false, secret: true }
+  ],
+
+  validate: async (config) => {
+    if (!config.url) return { ok: false, error: 'URL is required' };
+    if (!config.payloadTemplate) return { ok: false, error: 'Payload template is required' };
+    try {
+      if (config.headers) JSON.parse(config.headers);
+    } catch (e) {
+      return { ok: false, error: 'Headers must be valid JSON' };
+    }
+    return { ok: true };
+  },
+
+  sendMessage: async (config, payload) => {
+    const { to, message } = payload;
+    let payloadStr = config.payloadTemplate || '{}';
+    
+    // Replace template variables safely
+    payloadStr = payloadStr.replace(/\{\{to\}\}/g, to).replace(/\{\{message\}\}/g, message);
+
+    // If there is an API secret, replace {{secret}} in the headers or url
+    let headersObj = {};
+    if (config.headers) {
+      try {
+        const parsed = JSON.parse(config.headers);
+        for (const [k, v] of Object.entries(parsed)) {
+          headersObj[k] = config.apiSecret ? String(v).replace(/\{\{secret\}\}/g, config.apiSecret) : v;
+        }
+      } catch(e) {}
+    }
+
+    if (!headersObj['Content-Type']) {
+      headersObj['Content-Type'] = 'application/json';
+    }
+
+    const urlObj = new URL(config.url);
+    const options = {
+      hostname: urlObj.hostname,
+      port: urlObj.port || (urlObj.protocol === 'https:' ? 443 : 80),
+      path: urlObj.pathname + urlObj.search,
+      method: config.method || 'POST',
+      headers: headersObj
+    };
+
+    const client = urlObj.protocol === 'https:' ? https : http;
+
+    return new Promise((resolve, reject) => {
+      const req = client.request(options, (res) => {
+        let body = '';
+        res.on('data', chunk => body += chunk);
+        res.on('end', () => {
+          if (res.statusCode >= 200 && res.statusCode < 300) {
+            resolve({ status: res.statusCode, body });
+          } else {
+            reject(new Error(`Webhook failed: ${res.statusCode} ${body}`));
+          }
+        });
+      });
+      req.on('error', reject);
+      req.write(payloadStr);
+      req.end();
+    });
+  }
+};

nodejs/public/css/styles.css

@@ -3,10 +3,22 @@ nav.navbar{
 	padding-right: 1em;
 }
 
+/* Only the active top-nav link is bold + underlined; the username is plain. */
+.top-nav a.active{
+	font-weight: bold;
+	text-decoration: underline;
+}
+
 body {
 	display: flex;
 	flex-direction: column;
 	min-height: 100vh;
+	/* Height of the fixed navbar (plus the update banner, while shown --
+	   see top.ejs's showUpdateBanner/dismissUpdateBanner). Lets an in-page
+	   sticky element offset itself below both fixed elements via
+	   `top: var(--sw-content-offset)` instead of colliding with them at the
+	   viewport's true top:0. */
+	--sw-content-offset: 4.5rem;
 }
 
 #spa-shell {

nodejs/public/js/app.js

@@ -67,13 +67,6 @@ app.user = (function(app){
 		});
 	}
 
-	function remove(args, callack){
-		if(!confirm('Delete '+ args.uid+ 'user?')) return false;
-		app.api.delete('user/'+ args.uid, function(error, data){
-			callack(error, data);
-		});
-	}
-
 	function changePassword(args, callack){
 		app.api.put('users/'+ arg.uid || '', args, function(error, data){
 			callack(error, data);
@@ -102,7 +95,15 @@ app.user = (function(app){
 		});
 	}
 
-	return {list, remove, createInvite, setActive};
+	// A user DN's cn is always their uid (see models/user_ldap.js addLdapUser,
+	// `data.cn = data.uid`) -- pulling it straight out of the DN avoids an
+	// extra lookup just to display a manager list.
+	function dnToUid(dn){
+		var m = /^cn=([^,]+)/i.exec(dn || '');
+		return m ? m[1] : dn;
+	}
+
+	return {list, createInvite, setActive, dnToUid};
 
 })(app);
 
@@ -149,6 +150,21 @@ app.ui = (function(app){
 	// Drop the cache (e.g. after a group is created) so the next selector refetches.
 	function refreshGroups(){ _groupsPromise = null; return loadGroups(); }
 
+	// All usernames, fetched once and shared across every user selector (e.g. manager pickers).
+	var _usersPromise = null;
+	function loadUsers(){
+		if(!_usersPromise){
+			_usersPromise = new Promise(function(resolve){
+				app.user.list(function(error, data){
+					if(error || !data || !data.results){ resolve([]); return; }
+					resolve(data.results.map(function(u){ return u.uid; }).filter(Boolean).sort());
+				});
+			});
+		}
+		return _usersPromise;
+	}
+	function refreshUsers(){ _usersPromise = null; return loadUsers(); }
+
 	// opts: { values, options, freeSolo, placeholder, name, separator }
 	// Returns a handle: { get, set, add, clear, setOptions, element }.
 	function tagInput(mount, opts){
@@ -249,7 +265,25 @@ app.ui = (function(app){
 		return handle;
 	}
 
-	return { tagInput: tagInput, groupSelect: groupSelect, loadGroups: loadGroups, refreshGroups: refreshGroups };
+	// Universal user selector (e.g. picking managers). Preloads all usernames.
+	function userSelect(mount, opts){
+		opts = opts || {};
+		var handle = tagInput(mount, {
+			name: opts.name || 'manager',
+			values: opts.values || [],
+			options: [],
+			freeSolo: opts.freeSolo !== false,
+			separator: opts.separator != null ? opts.separator : '\n',
+			placeholder: opts.placeholder || 'Type a username…',
+		});
+		loadUsers().then(function(users){ handle.setOptions(users); });
+		return handle;
+	}
+
+	return {
+		tagInput: tagInput, groupSelect: groupSelect, loadGroups: loadGroups, refreshGroups: refreshGroups,
+		userSelect: userSelect, loadUsers: loadUsers, refreshUsers: refreshUsers,
+	};
 })(app);
 
 app.oauthClient = (function(app){
@@ -265,13 +299,6 @@ app.oauthClient = (function(app){
 		});
 	}
 
-	function remove(args, callack){
-		if(!confirm('Delete OAuth client "' + args.client_id + '"?')) return false;
-		app.api.delete('oauth/client/' + args.client_id, function(error, data){
-			callack(error, data);
-		});
-	}
-
 	function update(args, callack){
 		app.api.put('oauth/client/' + args.client_id, args, function(error, data){
 			callack(error, data);
@@ -284,7 +311,7 @@ app.oauthClient = (function(app){
 		});
 	}
 
-	return { list, add, remove, update, rotateSecret };
+	return { list, add, update, rotateSecret };
 })(app);
 
 app.tos = (function(app){
@@ -355,7 +382,7 @@ app.impersonate = (function(app){
 
 app.token = (function(app){
 	function list(name, callack){
-		if($.isFunction(name)){
+		if(typeof name === 'function'){
 			callack = name;
 			name = '';
 		}

nodejs/public/lib/js/app-base.js

@@ -1,3 +1,12 @@
+// Shared client framework for the theta42 apps.
+//
+// This file is byte-identical across sso-manager-node, proxy and jump-host —
+// per-app behaviour comes from the server (the `ui` locals in views/top.ejs and
+// the /api/user/me response), never from edits to this file. Edit all three
+// copies together.
+//
+// jQuery 4 safe: no $.isFunction, no $.holdReady.
+
 var app = {};
 
 app.pubsub = (function(){
@@ -45,7 +54,7 @@ app.pubsub = (function(){
 app.socket = (function(app){
 	// $.getScript('/socket.io/socket.io.js')
 	// <script type="text/javascript" src="/socket.io/socket.io.js"></script>
-	
+
 	var socket;
 	$(document).ready(function(){
 		socket = io({
@@ -75,11 +84,17 @@ app.socket = (function(app){
 app.api = (function(app){
 	var baseURL = '/api/'
 
-	function post(url, data, callback){
-		if (!$.isFunction(callback)) {
-			return new Promise((resolve, reject) => {
+	// post/put/delete are dual-mode: pass a callback for the node-style
+	// (error, data, status) form, or omit it to get a Promise that resolves
+	// with the parsed body and rejects with the error body. get/options return
+	// the jqXHR, which is itself thenable, so `await app.api.get(...)` works.
+
+	function body(method, url, data, callback){
+		if(typeof callback !== 'function'){
+			return new Promise(function(resolve, reject){
 				$.ajax({
-					type: 'POST', url: baseURL+url,
+					type: method,
+					url: baseURL+url,
 					headers: { 'auth-token': app.auth.getToken() },
 					data: JSON.stringify(data),
 					contentType: 'application/json; charset=utf-8',
@@ -88,9 +103,11 @@ app.api = (function(app){
 			});
 		}
 		return $.ajax({
-			type: 'POST',
+			type: method,
 			url: baseURL+url,
-			headers:{ 'auth-token': app.auth.getToken() },
+			headers:{
+				'auth-token': app.auth.getToken()
+			},
 			data: JSON.stringify(data),
 			contentType: "application/json; charset=utf-8",
 			dataType: "json",
@@ -104,40 +121,27 @@ app.api = (function(app){
 		});
 	}
 
+	function post(url, data, callback){
+		return body('POST', url, data, callback);
+	}
+
 	function put(url, data, callback){
-		if (!$.isFunction(callback)) {
-			return new Promise((resolve, reject) => {
-				$.ajax({
-					type: 'PUT', url: baseURL+url,
-					headers: { 'auth-token': app.auth.getToken() },
-					data: JSON.stringify(data),
-					contentType: 'application/json; charset=utf-8',
-					dataType: 'json',
-				}).done(resolve).fail(function(xhr){ reject(xhr.responseJSON || {}); });
-			});
-		}
-		return $.ajax({
-			type: 'PUT',
-			url: baseURL+url,
-			headers:{ 'auth-token': app.auth.getToken() },
-			data: JSON.stringify(data),
-			contentType: "application/json; charset=utf-8",
-			dataType: "json",
-			complete: function(res, text){
-				callback(
-					text !== 'success' ? res.statusText : null,
-					JSON.parse(res.responseText),
-					res.status
-				);
-			}
-		});
+		return body('PUT', url, data, callback);
 	}
 
-	function remove(url, callback){
-		if (!$.isFunction(callback)) {
-			return new Promise((resolve, reject) => {
+	// Called both as (url, callback) and — from formAJAX, which always passes
+	// the serialized form as the second argument — as (url, data, callback).
+	// No request body is sent either way.
+	function remove(url, data, callback){
+		if(typeof data === 'function'){
+			callback = data;
+			data = undefined;
+		}
+		if(typeof callback !== 'function'){
+			return new Promise(function(resolve, reject){
 				$.ajax({
-					type: 'DELETE', url: baseURL+url,
+					type: 'DELETE',
+					url: baseURL+url,
 					headers: { 'auth-token': app.auth.getToken() },
 					contentType: 'application/json; charset=utf-8',
 					dataType: 'json',
@@ -147,7 +151,9 @@ app.api = (function(app){
 		return $.ajax({
 			type: 'DELETE',
 			url: baseURL+url,
-			headers:{ 'auth-token': app.auth.getToken() },
+			headers:{
+				'auth-token': app.auth.getToken()
+			},
 			contentType: "application/json; charset=utf-8",
 			dataType: "json",
 			complete: function(res, text){
@@ -202,7 +208,10 @@ app.api = (function(app){
 })(app)
 
 app.auth = (function(app){
-	var user = {};
+	// One in-flight/cached GET /api/user/me per page load. Every gating
+	// decision (nav items, per-view forceLogin, group-required elements) reads
+	// this same promise instead of re-fetching.
+	var userPromise = null;
 
 	function setToken(token){
 		localStorage.setItem('APIToken', token);
@@ -216,35 +225,70 @@ app.auth = (function(app){
 		try{
 			return await app.api.get('user/me');
 		}catch(error){
-			if(error?.status === 401) return null;
-			throw error
+			if(error && error.status === 401) return null;
+			throw error;
 		}
 	}
 
+	// Cached current user, or false when there's no token at all. Callers that
+	// need a fresh copy (after a login or a profile change) pass force.
+	function loadUser(force){
+		if(force || !userPromise){
+			userPromise = getToken() ? getUser() : Promise.resolve(null);
+			userPromise = userPromise.then(function(user){
+				app.auth.user = app.auth.perms = user || null;
+				return user;
+			});
+		}
+		return userPromise;
+	}
+
+	// The apps report group membership two ways: sso-manager-node returns LDAP
+	// DNs in `memberOf`, the OIDC clients return plain CNs in `groups`. Both
+	// normalise to a list of CNs. `isAdmin` (the clients' effective-rights flag)
+	// is exposed as a synthetic `admin` group so one gating model covers both.
+	function groupCNs(user){
+		var raw = (user && (user.memberOf || user.groups)) || [];
+		if(!Array.isArray(raw)) raw = [raw];
+		var names = raw.map(function(group){
+			return String(group).split(',')[0].replace(/^cn=/i, '');
+		});
+		if(user && user.isAdmin && names.indexOf('admin') === -1) names.push('admin');
+		return names;
+	}
+
 	async function memberOf(groupNameToFind, user){
-		try{
-			user = user || await app.auth.asyncUser;
-			groupNameToFind = Array.isArray(groupNameToFind) ? groupNameToFind : [groupNameToFind]
+		user = user || await loadUser();
+		if(!user) return false;
+		groupNameToFind = Array.isArray(groupNameToFind) ? groupNameToFind : [groupNameToFind];
 
-			for(let group of user.memberOf){
-				group = group.split(',ou=groups')[0].replace('cn=', '');
-				if(groupNameToFind.includes(group)) return true;
-			}
-
-			return false;
-
-		}catch(error){
-			throw(error);
-		}
+		return groupCNs(user).some(function(group){
+			return groupNameToFind.includes(group);
+		});
 	}
 
-	async function isLoggedIn(){
-		if(getToken()){
-			user = await app.auth.asyncUser;
-			return user;
-		}else{
-			return false;
+	// True when the logged-in user is a global admin (per user/me). Sync — only
+	// meaningful once isLoggedIn/forceLogin has resolved.
+	function isAdmin(){
+		return !!(app.auth.perms && app.auth.perms.isAdmin);
+	}
+
+	// Dual-mode: returns a Promise resolving to the user (or false), and calls
+	// an optional node-style callback with the same result.
+	function isLoggedIn(callback){
+		var promise = loadUser().then(function(user){
+			return user || false;
+		});
+
+		if(typeof callback === 'function'){
+			promise.then(function(user){
+				callback(null, user);
+			}, function(error){
+				callback(error, false);
+			});
 		}
+
+		return promise;
 	}
 
 	function logIn(args, callback){
@@ -252,62 +296,125 @@ app.auth = (function(app){
 			if(data.login){
 				setToken(data.token);
 			}
+			loadUser(true);
 			callback(error, !!data.token);
 		});
 	}
 
+	// Clears the session only — the caller decides where to go next (the nav's
+	// Log Out button uses ui.logoutRedirect).
 	function logOut(callback){
 		localStorage.removeItem('APIToken');
-		location.replace(`/login${location.href.replace(location.origin, '')}`);
-		callback();
+		userPromise = null;
+		app.auth.user = app.auth.perms = null;
+		if(typeof callback === 'function') callback();
 	}
 
+	// Constrain a redirect target to a same-origin absolute path. Rejects
+	// absolute URLs (open redirect), protocol-relative "//host" and "/\host",
+	// and non-path schemes like "javascript:" (XSS). Falls back to "/".
+	function safeInternalPath(path){
+		if(typeof path !== 'string' || path.charAt(0) !== '/'
+				|| path.charAt(1) === '/' || path.charAt(1) === '\\'){
+			return '/';
+		}
+		return path;
+	}
+
+	// Consume an app token handed back by the OIDC callback via the URL
+	// fragment (#token=…&redirect=…). Stores it, strips the fragment, and
+	// forwards to the intended page. Returns true if a token was consumed.
+	function consumeTokenFragment(){
+		if(!location.hash) return false;
+		var params = new URLSearchParams(location.hash.replace(/^#/, ''));
+		var token = params.get('token');
+		if(!token) return false;
+
+		setToken(token);
+		// redirect comes from the URL fragment (attacker-controllable); only
+		// allow a same-origin path so it can't become an open redirect / XSS.
+		var redirect = safeInternalPath(params.get('redirect') || '/');
+		// Drop the token from the address bar before navigating on.
+		history.replaceState(null, '', location.pathname + location.search);
+		window.location.href = redirect;
+		return true;
+	}
+
+	// Page-level gate. jQuery 4 removed $.holdReady, so an unauthenticated or
+	// unauthorised user is kept off the page by a redirect / an error panel
+	// rather than by pausing document ready.
+	//
+	// `requiredGroups` is a group CN or an OR-list of them; the synthetic
+	// `admin` group covers the OIDC clients' isAdmin flag.
 	async function forceLogin(requiredGroups){
-		$.holdReady(true);
-		if(!await app.auth.isLoggedIn()) app.auth.logOut(function(){});
+		var user = await loadUser();
+
+		if(!user){
+			logOut(function(){});
+			location.replace('/login?redirect=' + encodeURIComponent(
+				location.pathname + location.search
+			));
+			return false;
+		}
 
 		if(user.onboardingRequired && location.pathname !== '/onboarding'){
 			location.replace('/onboarding');
+			return false;
 		}
 
-		if(requiredGroups){
-			if(!await memberOf(requiredGroups)){
-				console.log("Does not have permission!!!")
-				app.util.actionMessage(
-					`<h1>
-						<i class="fa-solid fa-triangle-exclamation"></i>
-						<b>You do not have permission to be here.</b>
-						<i class="fa-solid fa-triangle-exclamation"></i>
-					</h1>`,
-					$('#spa-shell'),
-					'danger',
-				);
-				throw new Error("User does not have permission");
-			}
+		if(requiredGroups && !await memberOf(requiredGroups, user)){
+			app.messages.action(
+				`<h1>
+					<i class="fa-solid fa-triangle-exclamation"></i>
+					<b>You do not have permission to be here.</b>
+					<i class="fa-solid fa-triangle-exclamation"></i>
+				</h1>`,
+				$('#spa-shell'),
+				'danger',
+			);
+			throw new Error("User does not have permission");
 		}
 
-		$.holdReady(false);	
+		return user;
 	}
 
+	// Where to go after a successful login: the ?redirect= query param, or the
+	// legacy /login/<path> suffix form, constrained to a same-origin path. The
+	// suffix form keeps its query string — /login/oauth/authorize?client_id=…
+	// is how the OIDC provider sends an unauthenticated user through login.
 	function logInRedirect(){
-		window.location.href = location.href.replace(location.origin+'/login', '') || '/'
+		var params = new URLSearchParams(location.search);
+		var target = params.get('redirect')
+			|| location.href.replace(location.origin + '/login', '')
+			|| '/';
+		window.location.href = safeInternalPath(target);
 	}
 
 	return {
 		getToken: getToken,
 		setToken: setToken,
+		getUser: getUser,
+		loadUser: loadUser,
+		groupCNs: groupCNs,
+		memberOf: memberOf,
+		isAdmin: isAdmin,
 		isLoggedIn: isLoggedIn,
+		safeInternalPath: safeInternalPath,
+		consumeTokenFragment: consumeTokenFragment,
+		user: null,
+		perms: null,
 		logIn: logIn,
 		logOut: logOut,
 		forceLogin,
 		logInRedirect,
-		getUser,
-		memberOf,
 	}
 
 })(app);
-app.auth.asyncUser = app.auth.getUser();
 
+// Back-compat alias for views that awaited the cached user directly.
+Object.defineProperty(app.auth, 'asyncUser', {
+	get: function(){ return app.auth.loadUser(); },
+});
 
 app.user = (function(app){
 	function list(callback){
@@ -338,6 +445,72 @@ app.user = (function(app){
 
 })(app);
 
+// Local (app-managed) permissions and groups. Only the OIDC-client apps serve
+// these endpoints; the calls are inert elsewhere.
+app.permission = (function(app){
+	function list(callback){
+		app.api.get('permission/', function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function subjects(callback){
+		app.api.get('permission/subjects', function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function add(args, callback){
+		app.api.post('permission/', args, function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function remove(id, callback){
+		app.api.delete('permission/' + encodeURIComponent(id), function(error, data){
+			callback(error, data);
+		});
+	}
+
+	return {list, subjects, add, remove};
+
+})(app);
+
+app.group = (function(app){
+	function list(callback){
+		app.api.get('group/', function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function add(args, callback){
+		app.api.post('group/', args, function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function remove(name, callback){
+		app.api.delete('group/' + encodeURIComponent(name), function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function addMember(name, username, callback){
+		app.api.post('group/' + encodeURIComponent(name) + '/members', {username}, function(error, data){
+			callback(error, data);
+		});
+	}
+
+	function removeMember(name, username, callback){
+		app.api.delete('group/' + encodeURIComponent(name) + '/members/' + encodeURIComponent(username), function(error, data){
+			callback(error, data);
+		});
+	}
+
+	return {list, add, remove, addMember, removeMember};
+
+})(app);
+
 app.util = (function(app){
 
 	function getUrlParameter(name){
@@ -347,65 +520,15 @@ app.util = (function(app){
 		return results === null ? '' : decodeURIComponent(results[1].replace(/\+/g, ' '));
 	};
 
-	function actionMessage(message, $targetPassed, type, callback){
-		message = message || '';
-
-		let $target = $targetPassed.closest('div.card').find('.actionMessage');
-		if(!$target.length) $target = $($targetPassed.find('.actionMessage')[0]);
-
-		type = type || 'info';
-		callback = callback || function(){};
-
-		if($target.html() === message) return;
-
-		if($target.html()){
-			$target.slideUp('fast', function(){
-				$target.html('')
-				$target.removeClass (function(index, className){
-					return (className.match (/(^|\s)bg-\S+/g) || []).join(' ');
-				});
-				if(message) return actionMessage(message, $target, type, callback);
-				$target.hide()
-			})
-		}else{
-			if(type) $target.addClass('bg-' + type);
-
-			if(!message.includes('<button')) message += `
-				<button class="action-close btn btn-sm btn-outline-dark float-end">
-					<i class="fa-solid fa-xmark"></i>
-				</button>
-			`
-			$target.html(message).slideDown('fast');
-		}
-		setTimeout(callback,10)
-	}
-
-	function actionConfirm(message, $target, type, callback){
-		return new Promise((resolve, reject) =>{
-			let id = crypto.randomUUID();
-			message = `
-				<h4 class"align-middle" >
-					<i class="fa-solid fa-triangle-exclamation"></i>
-					<b>${message}</b>
-					<span class="float-end">
-						<button type="button" class="btn btn-success confirm-${id}" data-confirm="true">
-							<i class="fa-solid fa-circle-check"></i>
-							Confirm
-						</button>
-						<button type="button" class="btn btn-danger confirm-${id}">
-							<i class="fa-solid fa-circle-stop"></i>
-							Cancel
-						</button>
-					</span>
-				</h4>
-			`
-			actionMessage(message, $target, type);
-			$("body").on('click', `.confirm-${id}`, function(){
-				actionMessage('', $target, type);
-				resolve(!!$(this).data('confirm'));
-			});
-		});
-
+	// escapeHtml/actionMessage/actionConfirm moved to @simpleworkjs/frontend's
+	// app.util.escapeHtml and app.messages.action/confirm.
+	function escapeHtml(s){
+		return String(s == null ? '' : s)
+			.replace(/&/g, '&amp;')
+			.replace(/</g, '&lt;')
+			.replace(/>/g, '&gt;')
+			.replace(/"/g, '&quot;')
+			.replace(/'/g, '&#39;');
 	}
 
 	$.fn.serializeObject = function() {
@@ -413,10 +536,12 @@ app.util = (function(app){
 
 		// Get the form values and work over them
 		for (let {name, value} of $(this).serializeArray()) {
-			console.log(name, value)
 			if (obj[name] === undefined) {
-				if (!value 
+				if (!value
 					&& !$(this).parent().find(`[name="${name}"]`).attr('value')
+					// Keep empty <textarea>s so a cleared field is submitted (and
+					// can reset a list, e.g. the per-host IP/header controls).
+					&& !$(this).filter(`textarea[name="${name}"]`).length
 				){
 					continue;
 				}
@@ -458,30 +583,79 @@ app.util = (function(app){
 		document.body.removeChild(element);
 	}
 
+	// Scroll a just-added/-edited element into view and flash its
+	// background, so the user's eye lands on the row that changed instead of
+	// it silently appearing/updating somewhere off-screen. Takes a jQuery
+	// object or a raw DOM node (e.g. jq-repeat's `item.__jq_$el`).
+	function revealItem(el){
+		var node = el && el.jquery ? el[0] : el;
+		if (!node) return;
+		if (typeof node.scrollIntoView === 'function') {
+			node.scrollIntoView({behavior: 'smooth', block: 'center'});
+		}
+		var prevTransition = node.style.transition;
+		var prevBg = node.style.backgroundColor;
+		node.style.transition = 'background-color 1.5s ease';
+		node.style.backgroundColor = 'var(--bs-success-bg-subtle, #d1e7dd)';
+		setTimeout(function(){
+			node.style.backgroundColor = prevBg;
+			setTimeout(function(){ node.style.transition = prevTransition; }, 1500);
+		}, 300);
+	}
+
 	return {
 		downloadFile: downloadFile,
 		getUrlParameter: getUrlParameter,
-		actionMessage: actionMessage,
-		actionConfirm,
+		escapeHtml: escapeHtml,
+		revealItem: revealItem,
 	}
 })(app);
 
-$( document ).ready(async function(){
+// Reveal every .group-required-<cn> element the current user's groups entitle
+// them to. Elements carrying .group-required start hidden (styles.css), so a
+// user who is in no groups — or who isn't logged in — simply never sees them.
+// The synthetic 'login' group is special: it's true for any authenticated user.
+app.auth.applyGroupVisibility = function(user){
+	var groups = app.auth.groupCNs(user);
+	var isLoggedIn = !!user;
 
-	// Show content if the user has the correct group
-	for(let group of (await app.auth.asyncUser)?.memberOf || []){
+	var style = document.getElementById('group-required-rules');
+	if(!style){
+		style = document.createElement('style');
+		style.id = 'group-required-rules';
+		document.head.appendChild(style);
+	}
+
+	for(var group of groups){
 		try{
-			group = group.split(',ou=groups')[0].replace('cn=', '');
-
-			const sheet = document.styleSheets[0];
-			const selector = `.group-required-${group}`;
-			const cssText = `${selector} { display: revert !important; }`;
-			sheet.insertRule(cssText, sheet.cssRules.length);
+			style.sheet.insertRule(
+				`.group-required-${CSS.escape(group)} { display: revert !important; }`,
+				style.sheet.cssRules.length
+			);
 		}catch(error){
-			
+			// A group whose CN isn't a usable CSS identifier just gates nothing.
 		}
 	}
 
+	// The 'login' group is synthetic — it means "any authenticated user".
+	// Reveal .group-required-login for any logged-in user.
+	if(isLoggedIn){
+		try{
+			style.sheet.insertRule(
+				`.group-required-login { display: revert !important; }`,
+				style.sheet.cssRules.length
+			);
+		}catch(error){
+			// Ignore CSS escape errors.
+		}
+	}
+};
+
+$( document ).ready(async function(){
+
+	// Show content the user's groups entitle them to.
+	app.auth.applyGroupVisibility(await app.auth.loadUser());
+
 	$('div.row').fadeIn('slow'); //show the page
 
 	//panel button's
@@ -502,9 +676,9 @@ $( document ).ready(async function(){
 		$(this).closest('.card').slideUp('fast');
 	});
 
-	$('.actionMessage').on('click', 'button.action-close', function(event){
-		app.util.actionMessage(null, $(this));
-	});
+	// action-close click handling is wired by @simpleworkjs/frontend's
+	// app.messages.js (delegated on document, so it also covers messages
+	// rendered after this ready handler runs).
 
 	setInterval(()=>{
 		$('.momentFromNow').each((idx, el)=>{
@@ -521,7 +695,6 @@ $( document ).ready(async function(){
 		const yOffset = Number($('#spa-shell').css('margin-top').replace('px', ''));
 		const y = this[0].getBoundingClientRect().top + window.scrollY - yOffset;
 
-		console.log('y', y)
 		window.scrollTo({top: y, behavior: 'smooth'});
 	};
 
@@ -535,31 +708,26 @@ function formAJAX(btn){
 	var method = ($form.attr('method') || 'post').toLowerCase();
 
 	if($form.validate && !$form.validate()){
-		app.util.actionMessage('Please fix the form errors.', $form, 'danger')
+		app.messages.action('Please fix the form errors.', $form, 'danger')
 		return false;
 	}
-	
-	app.util.actionMessage( 
-		`<div class="spinner-border" role="status">
-			<span class="visually-hidden">Loading...</span>
-		</div>`,
-		$form,
-		'info'
-	);
+
+	// Plain text: app.messages.action HTML-escapes its message (by design,
+	// see @simpleworkjs/frontend), so raw markup like a spinner <div> would
+	// render literally instead of as an element.
+	app.messages.action('Saving…', $form, 'info');
 
 	app.api[method]($form.attr('action'), formData, function(error, data){
-		app.util.actionMessage(data.message, $form, error ? 'danger' : 'success'); //re-populate table
+		app.messages.action(data.message, $form, error ? 'danger' : 'success'); //re-populate table
 		$form.validateClear();
 		if(!error){
 			$form.trigger("reset");
 			eval($form.attr('evalAJAX')); //gets JS to run after completion
 		}else{
-			console.log('formAJAX res error', error, data)
 			if(data && data.name === 'ObjectValidateError'){
-				app.util.actionMessage('Please fix the form errors', $form, 'danger'); //re-populate table
+				app.messages.action('Please fix the form errors', $form, 'danger'); //re-populate table
 			}
 			if(data && data.keys){
-				console.log('form key errors', data.keys)
 				for(let keyError of data.keys){
 					$form.find(`[name=${keyError.key}]`).validateMessage(keyError.message);
 				}
@@ -567,4 +735,3 @@ function formAJAX(btn){
 		}
 	});
 }
-

nodejs/public/lib/js/val.js

@@ -1,133 +0,0 @@
-( function( $ ) {
-	var settings = {
-		rule: {
-			eq: function(value, options){
-				var compare = $('[name=' + options + ']').val();
-	 
-				if ( value != compare ) {
-					return "Miss-match";
-				}
-			}
-		},
-	};
-
-	$.fn.validate = function(event) {
-		// let	thisSettings = $.extend(true, settings, settingsObj);
-		let hasErrors = false;
-
-		if(this.is('[validate]')) return this.validateField(event);
-
-		if(!this.attr('isValid')){
-			console.log('adding reset event')
-			this.on('reset', function(){
-				$(this).attr('isValid', false);
-				$(this).validateClear();
-			})
-		}
-
-		this.find('[validate]').each(function(){
-			if(!$(this).validateField()) hasErrors = true;
-		});
-		
-		this.attr('isValid', !hasErrors);
-
-		if(hasErrors && event) event.preventDefault();
-
-		return !hasErrors;
-	};
-
-	$.fn.validateClear = function(){
-		$(this).find('input').each(function(){
-			$(this).removeClass('is-invalid');
-			$(this).removeClass('is-valid');
-		})
-	}
-
-	$.fn.validateField = function(){
-		var attr = this.attr('validate').split(':'); //array of params
-		var	rule = attr[0];
-		var	options = attr[1];
-		var	value = this.val(); //link to input value
-		var message;
-
-		if(this.prop('disabled')) return true;
-
-
-		//checks if field is required, and length 
-		if(!isNaN(options) && value.length < options){
-			message = `Must be ${options} characters`;
-		}
-
-		//checks if empty to stop processing 
-		if(!isNaN(options) && value.length === 0) {
-		}else if(rule in settings.rule){
-			let message = settings.rule[rule].apply(this, [value, options]);
-		}
-
-		this.validateMessage(message)
-		return !message;
-	}
-
-	$.fn.validateMessage = function(message){
-		if(message && message !== true){
-			this.closest('.form-group').find('b.invalid-feedback').html(message);
-			this.addClass('is-invalid');
-		}else{
-			this.removeClass('is-invalid');
-			this.addClass('is-valid');
-		}
-		return this;
-	};
-
-	jQuery.extend({
-		validateSettings: function( settingsObj ) {
-			$.extend( true, settings, settingsObj );
-		},
-
-		validateInit: function( ettingsObj ) {
-			$( '[action]' ).on( 'submit', function ( event, settingsObj ){
-				$( this ).validate( settingsObj, event );
-			});
-		}
-	});
-
-}( jQuery ));
-
-$.validateSettings({
-	rule:{
-		ip: function( value ) {
-			value = value.split( '.' );
- 
-			if ( value.length != 4 ) {
-				return "Malformed IP";
-			}
- 
-			$.each( value, function( key, value ) {
-				if( value > 255 || value < 0 ) {
-					return "Malformed IP";
-				}
-			});
-		},
-
-		host: function( value ) {
-			var reg = /^(?=.{1,255}$)[0-9A-Za-z](?:(?:[0-9A-Za-z]|-){0,61}[0-9A-Za-z])?(?:\.[0-9A-Za-z](?:(?:[0-9A-Za-z]|-){0,61}[0-9A-Za-z])?)*\.?$/;
-			if ( reg.test( value ) === false ) {
-				return "Invalid";
-			}
-		},
-
-		user: function( value ) {
-			var reg = /^[a-z0-9\_\-\@\.]{1,32}$/;
-			if ( reg.test( value ) === false ) {
-				return "Invalid";
-			}
-		},
- 
-		password: function( value ) {
-			var reg = /^(?=[^\d_].*?\d)\w(\w|[!@#$%]){1,48}/;
-			if ( reg.test( value ) === false ) {
-				return "Weak password, Try again";
-			}
-		}
-	}
-});

nodejs/public/resources/theta-agent/install.sh

@@ -0,0 +1,247 @@
+#!/bin/bash
+set -e
+
+# --- Configuration ---
+# In a real environment, these would be derived from the script's download URL
+# or passed as additional arguments. For now, we use the most recent release.
+BINARY_URL="https://github.com/theta42/theta-agent/releases/latest/download/theta-agent-linux-amd64"
+CONFIG_DIR="/etc/theta42"
+CONFIG_FILE="$CONFIG_DIR/agent.yml"
+BIN_PATH="/usr/local/bin/theta-agent"
+SERVICE_FILE="/etc/systemd/system/theta-agent.service"
+
+# Colors for output
+RED='\033[0;31m'
+GREEN='\033[0;32m'
+NC='\033[0m' # No Color
+
+log() { echo -e "${GREEN}[+]${NC} $1"; }
+error() { echo -e "${RED}[!]${NC} $1"; exit 1; }
+
+# 1. Root check
+if [ "$(id -u 2>/dev/null || echo 1)" -ne 0 ]; then
+  error "This script must be run as root."
+fi
+
+# Install SSSD and PAM integration packages if missing
+install_sssd_deps() {
+  if ! command -v sssd >/dev/null 2>&1; then
+    log "Installing SSSD and PAM integration dependencies..."
+    if command -v apt-get >/dev/null 2>&1; then
+      DEBIAN_FRONTEND=noninteractive apt-get update -qq || true
+      DEBIAN_FRONTEND=noninteractive apt-get install -y -qq sssd sssd-ldap libnss-sss libpam-sss libsss-sudo libpam-runtime || \
+      DEBIAN_FRONTEND=noninteractive apt-get install -y -qq sssd sssd-ldap libnss-sss libpam-sss || true
+      if command -v pam-auth-update >/dev/null 2>&1; then
+        pam-auth-update --package --enable mkhomedir sss || pam-auth-update --enable mkhomedir || true
+      fi
+    elif command -v dnf >/dev/null 2>&1; then
+      dnf install -y sssd sssd-ldap sssd-tools || true
+    elif command -v yum >/dev/null 2>&1; then
+      yum install -y sssd sssd-ldap sssd-tools || true
+    elif command -v pacman >/dev/null 2>&1; then
+      pacman -S --noconfirm sssd || true
+    elif command -v zypper >/dev/null 2>&1; then
+      zypper in -y sssd || true
+    fi
+  else
+    log "SSSD is already installed."
+  fi
+  mkdir -p /etc/sssd
+  chmod 755 /etc/sssd
+}
+
+# 2. Argument Parsing
+URL=""
+TOKEN=""
+JOIN_KEY=""
+PUBLIC_KEY=""
+B64_CONFIG=""
+INSTALL_SSSD=0
+
+while [ $# -gt 0 ]; do
+  case $1 in
+    --url)
+      URL="$2"
+      shift 2
+      ;;
+    --token)
+      TOKEN="$2"
+      shift 2
+      ;;
+    # Base64 of the SSO's raw Ed25519 public key. The agent verifies high-risk
+    # commands (reboot, configure_ldap, arbitrary_bash, update_binary) against
+    # it and REFUSES them when it is absent, so an install without this key can
+    # stream telemetry but cannot be acted on.
+    --public-key)
+      PUBLIC_KEY="$2"
+      shift 2
+      ;;
+    # The one credential an operator hands out. The server exchanges it for a
+    # per-agent token on first connect, which the agent writes back into
+    # agent.yml -- so this is all you need to add a host.
+    --join-key)
+      JOIN_KEY="$2"
+      shift 2
+      ;;
+    --install-sssd|--ldap)
+      INSTALL_SSSD=1
+      shift
+      ;;
+    *)
+      B64_CONFIG="$1"
+      shift
+      ;;
+  esac
+done
+
+# Validation: require credentials ONLY if config file does not already exist
+if [ ! -f "$CONFIG_FILE" ] && [ -z "$B64_CONFIG" ] && { [ -z "$URL" ] || { [ -z "$TOKEN" ] && [ -z "$JOIN_KEY" ]; }; }; then
+  error "Missing required configuration. Provide a base64 encoded config, or --url with either --join-key or --token."
+  echo "Usage examples:"
+  echo "  sh install.sh \"BASE64_CONFIG\""
+  echo "  sh install.sh --url \"https://sso.local\" --join-key \"tjk_...\" --install-sssd"
+  echo "  sh install.sh --url \"https://sso.local\" --token \"ISSUED_TOKEN\" --public-key \"BASE64_KEY\""
+  echo ""
+  echo "--join-key is the normal path: the host enrolls itself on first connect"
+  echo "and the SSO issues it its own token + public key, which the agent writes"
+  echo "back into agent.yml. Get a key from Directory -> Install Agent."
+  exit 1
+fi
+
+log "Starting Theta Agent installation..."
+
+# Architecture and OS detection
+OS_NAME="$(uname -s | tr '[:upper:]' '[:lower:]')"
+ARCH_NAME="$(uname -m)"
+BINARY_NAME="theta-agent-linux-amd64"
+
+case "$OS_NAME" in
+  linux*)
+    case "$ARCH_NAME" in
+      x86_64|amd64) BINARY_NAME="theta-agent-linux-amd64" ;;
+      aarch64|arm64) BINARY_NAME="theta-agent-linux-arm64" ;;
+      armv7*|armhf) BINARY_NAME="theta-agent-linux-armv7" ;;
+      *) BINARY_NAME="theta-agent-linux-amd64" ;;
+    esac
+    ;;
+  darwin*)
+    case "$ARCH_NAME" in
+      x86_64|amd64) BINARY_NAME="theta-agent-darwin-amd64" ;;
+      arm64|aarch64) BINARY_NAME="theta-agent-darwin-arm64" ;;
+      *) BINARY_NAME="theta-agent-darwin-arm64" ;;
+    esac
+    ;;
+  mingw*|msys*|cygwin*)
+    case "$ARCH_NAME" in
+      aarch64|arm64) BINARY_NAME="theta-agent-windows-arm64.exe" ;;
+      *) BINARY_NAME="theta-agent-windows-amd64.exe" ;;
+    esac
+    ;;
+esac
+
+BINARY_URL="https://github.com/theta42/theta-agent/releases/latest/download/${BINARY_NAME}"
+
+# 3. Install binary
+log "Detected OS: $OS_NAME ($ARCH_NAME) -> Downloading binary $BINARY_NAME..."
+curl -fsSL "$BINARY_URL" -o "$BIN_PATH.tmp" || error "Failed to download binary from $BINARY_URL"
+chmod +x "$BIN_PATH.tmp"
+mv -f "$BIN_PATH.tmp" "$BIN_PATH"
+
+# 4. Setup configuration
+log "Preparing configuration directory $CONFIG_DIR..."
+mkdir -p "$CONFIG_DIR"
+chmod 755 "$CONFIG_DIR"
+
+if [ -n "$B64_CONFIG" ]; then
+  log "Decoding and writing configuration from base64..."
+  echo "$B64_CONFIG" | base64 -d > "$CONFIG_FILE" || error "Failed to decode base64 configuration."
+elif [ ! -f "$CONFIG_FILE" ]; then
+  log "Generating minimal configuration from arguments..."
+  cat <<EOF > "$CONFIG_FILE"
+server_url: "$URL"
+auth_token: "$TOKEN"
+join_key: "$JOIN_KEY"
+public_key: "$PUBLIC_KEY"
+location: "unknown"
+capabilities:
+  telemetry: true
+  configure_ldap: true
+  ldap_tunnel: true
+  reboot: false
+  service_control: []
+  arbitrary_bash: false
+EOF
+else
+  log "Preserving existing configuration at $CONFIG_FILE"
+fi
+# Ensure theta-secrets & theta groups exist for non-root secret access
+log "Configuring non-root secret access groups (theta-secrets)..."
+if command -v groupadd >/dev/null 2>&1; then
+  getent group theta-secrets >/dev/null 2>&1 || groupadd -r theta-secrets 2>/dev/null || true
+  getent group theta >/dev/null 2>&1 || groupadd -r theta 2>/dev/null || true
+fi
+SECRETS_GROUP="root"
+if getent group theta-secrets >/dev/null 2>&1; then
+  SECRETS_GROUP="theta-secrets"
+elif getent group theta >/dev/null 2>&1; then
+  SECRETS_GROUP="theta"
+fi
+chown -R "root:$SECRETS_GROUP" "$CONFIG_DIR" 2>/dev/null || true
+chmod 750 "$CONFIG_DIR"
+chmod 640 "$CONFIG_FILE"
+
+# 4c. Setup Desktop Tray Icon companion
+TRAY_BINARY_NAME="theta-agent-tray-${OS_NAME}-${ARCH_NAME}"
+case "$OS_NAME" in
+  linux*) TRAY_BINARY_NAME="theta-agent-tray-linux-amd64" ;;
+  windows*) TRAY_BINARY_NAME="theta-agent-tray-windows-amd64.exe" ;;
+esac
+TRAY_BIN_PATH="/usr/local/bin/theta-agent-tray"
+TRAY_URL="https://github.com/theta42/theta-agent/releases/latest/download/${TRAY_BINARY_NAME}"
+
+log "Attempting to install desktop tray companion ($TRAY_BINARY_NAME)..."
+if curl -fsSL "$TRAY_URL" -o "$TRAY_BIN_PATH.tmp" 2>/dev/null; then
+  chmod +x "$TRAY_BIN_PATH.tmp"
+  mv -f "$TRAY_BIN_PATH.tmp" "$TRAY_BIN_PATH"
+  mkdir -p /etc/xdg/autostart
+  cat <<EOF > /etc/xdg/autostart/theta-agent-tray.desktop
+[Desktop Entry]
+Type=Application
+Name=Theta Agent Tray
+Comment=Theta Agent Desktop Tray Companion
+Exec=/usr/local/bin/theta-agent-tray
+Icon=network-workgroup
+Terminal=false
+Categories=Utility;System;
+X-GNOME-Autostart-enabled=true
+EOF
+  log "Desktop tray companion installed at $TRAY_BIN_PATH with autostart."
+fi
+
+# 5. Setup systemd service
+log "Creating systemd service unit..."
+cat <<EOF > "$SERVICE_FILE"
+[Unit]
+Description=Theta Agent Unified Endpoint Management
+After=network.target
+
+[Service]
+Type=simple
+ExecStart=$BIN_PATH
+Restart=always
+RestartSec=5
+SyslogIdentifier=theta-agent
+
+[Install]
+WantedBy=multi-user.target
+EOF
+
+# 6. Start the agent
+log "Enabling and starting Theta Agent..."
+systemctl daemon-reload
+systemctl enable theta-agent
+systemctl start theta-agent
+
+log "Theta Agent installation complete!"
+log "Verify status with: systemctl status theta-agent"
+log "Check logs with: journalctl -u theta-agent -f"

nodejs/routes/access_request.js

@@ -0,0 +1,266 @@
+'use strict';
+
+// Self-service access requests. Mounted at /api/access-requests (app.js).
+//
+// The loop this closes: a user browses the catalog, finds something they cannot
+// reach, asks for it; the resource's owner (or a directory admin) approves; the
+// approval performs the LDAP group add. LDAP stays the access-control truth --
+// this router never invents a permission, it only automates the group add an
+// admin would otherwise do by hand, and records who decided.
+
+const router = require('express').Router();
+const { Resource, ResourceGroup } = require('../models/resource');
+const { AccessRequest, STATUS } = require('../models/access_request');
+const { Group } = require('../models/group_ldap');
+const { User } = require('../models/user_ldap');
+const { Mail } = require('../models/email');
+const { groupCns } = require('../utils/user_groups');
+const { envelope, projectResource } = require('@simpleworkjs/directory-schema');
+
+const DIRECTORY_ADMIN_GROUPS = ['app_sso_directory_admin', 'app_sso_admin', 'app_super_admin'];
+
+function httpError(status, message) {
+  const err = new Error(message);
+  err.status = status;
+  return err;
+}
+
+// May `user` decide requests against `resource`? The resource's own owner is
+// the primary approver -- that is the point of Resource.owner -- with directory
+// admins as the catch-all so an unowned or orphaned resource is never stuck.
+async function canDecide(user, resource, callerGroups) {
+  if (resource && resource.owner && resource.owner === user.uid) return true;
+  return callerGroups.some(g => DIRECTORY_ADMIN_GROUPS.includes(g));
+}
+
+// The group that satisfies a request for this resource. Prefers an explicit
+// choice, else the `member`-level link (the "just let me use it" group) over an
+// `owner`-level one -- requesting a resource should never silently escalate to
+// its admin group.
+async function resolveGroupCn(resourceId, requested) {
+  const links = await ResourceGroup.list({ where: { resourceId } });
+  if (!links.length) {
+    throw httpError(409, 'This resource has no access group linked, so it cannot be requested.');
+  }
+  if (requested) {
+    const match = links.find(l => l.groupCn === requested);
+    if (!match) throw httpError(400, `"${requested}" is not an access group for this resource.`);
+    return match.groupCn;
+  }
+  const member = links.find(l => l.accessLevel === 'member');
+  return (member || links[0]).groupCn;
+}
+
+// Best-effort notification. A mail failure must never fail the request itself --
+// the row is the source of truth and the approver can find it in the UI.
+async function notify(uid, subject, message) {
+  try {
+    const user = await User.get({ uid });
+    if (!user || !user.mail) return;
+    await Mail.sendTemplate(user.mail, 'notification', {
+      givenName: user.givenName || uid,
+      subject,
+      message,
+    });
+  } catch (err) {
+    console.error(`access-request: notification to ${uid} failed:`, err.message);
+  }
+}
+
+// POST /api/access-requests  { slug | resourceId, groupCn?, note? }
+router.post('/', async (req, res, next) => {
+  try {
+    if (req.user.isMachine) throw httpError(403, 'Machine accounts cannot request access.');
+
+    let resource;
+    if (req.body.slug) {
+      const found = await Resource.list({ where: { slug: req.body.slug } });
+      resource = found[0];
+    } else if (req.body.resourceId) {
+      resource = await Resource.get(req.body.resourceId);
+    }
+    if (!resource) throw httpError(404, 'Resource not found');
+
+    const md = resource.metadata || {};
+    // Opt-out, not opt-in: everything in the catalog is requestable unless an
+    // admin has explicitly marked it otherwise.
+    if (md.requestable === false) {
+      throw httpError(409, 'This resource is not available for self-service requests.');
+    }
+
+    const groupCn = await resolveGroupCn(resource.id, req.body.groupCn);
+
+    const callerGroups = await groupCns(req.user);
+    if (callerGroups.includes(groupCn)) {
+      throw httpError(409, 'You already have access to this resource.');
+    }
+
+    const existing = await AccessRequest.findOpen(req.user.uid, groupCn);
+    if (existing) throw httpError(409, 'You already have a pending request for this resource.');
+
+    const request = await AccessRequest.create({
+      uid: req.user.uid,
+      resourceId: resource.id,
+      groupCn,
+      status: STATUS.PENDING,
+      note: req.body.note || '',
+      requestedOn: Date.now(),
+    });
+
+    if (resource.owner) {
+      await notify(
+        resource.owner,
+        `Access request: ${resource.name}`,
+        `<p><strong>${req.user.uid}</strong> has requested access to <strong>${resource.name}</strong> (group <code>${groupCn}</code>).</p>` +
+        (req.body.note ? `<p>Their note: ${req.body.note}</p>` : '') +
+        `<p>Review it on the Directory page.</p>`
+      );
+    }
+
+    res.json(envelope(request));
+  } catch (err) { next(err); }
+});
+
+// GET /api/access-requests/mine — the caller's own request history.
+router.get('/mine', async (req, res, next) => {
+  try {
+    const rows = await AccessRequest.listForUser(req.user.uid);
+    res.json(envelope(await decorate(rows)));
+  } catch (err) { next(err); }
+});
+
+// GET /api/access-requests — pending requests the caller may decide.
+router.get('/', async (req, res, next) => {
+  try {
+    const callerGroups = await groupCns(req.user);
+    const isAdmin = callerGroups.some(g => DIRECTORY_ADMIN_GROUPS.includes(g));
+    const pending = await AccessRequest.listPending();
+
+    let visible = pending;
+    if (!isAdmin) {
+      // A plain resource owner sees only requests against resources they own.
+      const owned = await Resource.list({ where: { owner: req.user.uid } });
+      const ownedIds = new Set(owned.map(r => r.id));
+      visible = pending.filter(r => ownedIds.has(r.resourceId));
+    }
+    res.json(envelope(await decorate(visible)));
+  } catch (err) { next(err); }
+});
+
+// Attach the resource name/slug each row refers to. The UI needs it on every
+// list and would otherwise issue one lookup per row.
+async function decorate(rows) {
+  if (!rows.length) return [];
+  const resources = await Resource.list();
+  const byId = new Map(resources.map(r => [r.id, r]));
+  return rows.map(row => {
+    const data = row.toJSON ? row.toJSON() : { ...row };
+    const resource = byId.get(data.resourceId);
+    data.resource = resource
+      ? { id: resource.id, name: resource.name, slug: resource.slug, kind: resource.kind }
+      : null;
+    return data;
+  });
+}
+
+// POST /api/access-requests/:id/approve  { decisionNote? }
+router.post('/:id/approve', async (req, res, next) => {
+  try {
+    const request = await AccessRequest.get(req.params.id);
+    if (!request) throw httpError(404, 'Request not found');
+    if (request.status !== STATUS.PENDING) {
+      throw httpError(409, `This request was already ${request.status}.`);
+    }
+
+    const resource = await Resource.get(request.resourceId);
+    const callerGroups = await groupCns(req.user);
+    if (!(await canDecide(req.user, resource, callerGroups))) {
+      throw httpError(403, 'You do not have permission to decide this request.');
+    }
+
+    // The LDAP write happens FIRST and is allowed to throw. Marking a request
+    // approved without the group add would show the user a grant they do not
+    // actually have -- a pending row is recoverable, a lying one is not.
+    const group = await Group.get(request.groupCn);
+    const user = await User.get({ uid: request.uid });
+    try {
+      await group.addMember(user);
+    } catch (err) {
+      // "already a member" is the goal state, not a failure. This happens
+      // routinely: groupOfNames requires at least one member, so creating a
+      // resource seeds its auto-created groups with the creator's DN, and an
+      // admin may also grant access by hand while a request sits pending.
+      // Without this the request would 500 and stay pending forever.
+      const alreadyMember = err.name === 'TypeOrValueExistsError' || err.code === 20;
+      if (!alreadyMember) throw err;
+    }
+    User.clearCache(); // membership feeds cached isAdmin / group-gated nav
+
+    const updated = await request.update({
+      status: STATUS.APPROVED,
+      decidedBy: req.user.uid,
+      decidedOn: Date.now(),
+      decisionNote: req.body.decisionNote || '',
+    });
+
+    await notify(
+      request.uid,
+      `Access approved: ${resource ? resource.name : request.groupCn}`,
+      `<p>Your request for <strong>${resource ? resource.name : request.groupCn}</strong> was approved by ${req.user.uid}.</p>` +
+      `<p>You may need to sign out and back in for the change to take effect everywhere.</p>`
+    );
+
+    res.json(envelope(updated));
+  } catch (err) { next(err); }
+});
+
+// POST /api/access-requests/:id/deny  { decisionNote? }
+router.post('/:id/deny', async (req, res, next) => {
+  try {
+    const request = await AccessRequest.get(req.params.id);
+    if (!request) throw httpError(404, 'Request not found');
+    if (request.status !== STATUS.PENDING) {
+      throw httpError(409, `This request was already ${request.status}.`);
+    }
+
+    const resource = await Resource.get(request.resourceId);
+    const callerGroups = await groupCns(req.user);
+    if (!(await canDecide(req.user, resource, callerGroups))) {
+      throw httpError(403, 'You do not have permission to decide this request.');
+    }
+
+    const updated = await request.update({
+      status: STATUS.DENIED,
+      decidedBy: req.user.uid,
+      decidedOn: Date.now(),
+      decisionNote: req.body.decisionNote || '',
+    });
+
+    await notify(
+      request.uid,
+      `Access request declined: ${resource ? resource.name : request.groupCn}`,
+      `<p>Your request for <strong>${resource ? resource.name : request.groupCn}</strong> was declined.</p>` +
+      (req.body.decisionNote ? `<p>Reason: ${req.body.decisionNote}</p>` : '')
+    );
+
+    res.json(envelope(updated));
+  } catch (err) { next(err); }
+});
+
+// DELETE /api/access-requests/:id — requester withdraws their own pending request.
+router.delete('/:id', async (req, res, next) => {
+  try {
+    const request = await AccessRequest.get(req.params.id);
+    if (!request) throw httpError(404, 'Request not found');
+    if (request.uid !== req.user.uid) {
+      throw httpError(403, 'You can only withdraw your own requests.');
+    }
+    if (request.status !== STATUS.PENDING) {
+      throw httpError(409, `This request was already ${request.status}.`);
+    }
+    const updated = await request.update({ status: STATUS.CANCELLED, decidedOn: Date.now() });
+    res.json(envelope(updated));
+  } catch (err) { next(err); }
+});
+
+module.exports = router;