Compare commits

...
13 Commits
Author SHA1 Message Date
mrcharles 2440188f1f feat(tooling): containerize dns-cli in tooling-cli/dns
Consolidate the Technitium DNS CLI into tooling-cli/dns/ with a
containerized bash CLI (src/dns_cli.sh), Dockerfile (alpine +
curl/python3/bind-tools/bash), README, AGENTS.md, validate.sh, and
.env.example. Env vars aligned to the centralized ~/.creds/technitium.env
store (TECHNITIUM_URL, TECHNITIUM_DNS_TOKEN, TECHNITIUM_DNS_ZONE).

Image built and pushed to the registry as
git.knownelement.com/reachableceo/dns-cli:latest. Verified live:
zones, list, and search all return data.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-08-10 10:21:37 -05:00
mrcharles 67f8056e0b refactor(tooling): merge redmine-cli into tooling-cli/redmine
Consolidate the redmine-cli source into tooling-cli/redmine/
(CLI source, Dockerfile, README, AGENTS.md) with all documentation
rewritten to invoke the container via raw docker run and credentials
from the centralized ~/.creds/redmine.env store. No bin/ wrapper, no
system-dependent paths in the docs.

Removes the old redmine-cli/ subdirectory, updates the
KNELCredsManager consumer table, and updates STATUS.md + AGENTS.md
to reference the new location and registry image.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-08-10 09:53:42 -05:00
mrcharles ecffb4e598 refactor(tooling): merge discourse-cli into tooling-cli/discourse
Consolidate the discourse-cli source into tooling-cli/discourse/
(CLI source, Dockerfile, README, AGENTS.md, validate.sh) with all
documentation rewritten to invoke the container via raw docker run
and credentials from the centralized ~/.creds/discourse.env store.
No bin/ wrapper, no system-dependent paths in the docs.

Removes the old discourse-cli/ subdirectory and updates the
KNELCredsManager consumer table to reference the container directly.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-08-10 09:49:22 -05:00
mrcharles 1957dcbc7e feat(creds): centralize credentials to ~/.creds/, add KNELCredsManager
Rewire all MCP wrappers, docker-compose services, and validate scripts
to source credentials from ~/.creds/ instead of scattered per-service
.env files. This removes credential duplication and prepares for the
HashiCorp Vault migration.

Add KNELCredsManager under tooling-cli with:
- Containerized Bitwarden CLI (pinned image, host stays clean)
- scripts/bw wrapper with session management and data persistence
- README documenting credential layout, consumer wiring, and roadmap

Also fixes latent bug in MCP wrappers that were silently getting empty
creds from ambient shell env — they now explicitly source ~/.creds/.

Tracked in Redmine #407. Discourse: https://community.turnsys.com/t/308

💘 Generated with Crush

Assisted-by: Crush
2026-08-10 09:14:14 -05:00
mrcharles 7dfce930b6 fix: resolve subcategory slugs in category lookup and listing
The /categories.json endpoint only returns top-level categories by
default, so subcategory slugs (e.g. vp-techops) failed to resolve and
the 'categories' listing hid the entire subcategory tree. Both
_resolve_category and cmd_categories now pass include_subcategories=true
and traverse the nested subcategory_list.
2026-08-06 14:13:13 -05:00
mrcharles 11ec4cf4a3 fix: resolve discourse-mcp TS2345 build error and add env-var auth
The discourse-mcp was BLOCKED with a TypeScript TS2345 error caused by
the Dockerfile forcing SDK ^1.23.0 via `pnpm add`, while the upstream
code requires the lockfile's v1.17.3. Removed the override to use
`pnpm install --frozen-lockfile`.

The upstream server does not read environment variables directly — it
requires CLI args (--auth-pairs, --site) or a --profile JSON file. Added
a docker-entrypoint.sh that converts DISCOURSE_URL, DISCOURSE_API_KEY,
and DISCOURSE_API_USERNAME env vars into a profile JSON with auth_pairs
and site tethering.

Validated end-to-end: MCP handshake passes, site auto-tethers on startup,
live tool calls (discourse_search, discourse_filter_topics) return real
data from community.turnsys.com.

Updates STATUS.md and README.md to reflect discourse-mcp as production
ready. Adds JOURNAL.md entries for discourse-cli, discourse-mcp fix, and
protocol improvements.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-07-30 16:53:12 -05:00
mrcharles e2d4bb333c fix: enforce one-server-at-a-time workflow in agent protocol
Agents were attempting to build/validate multiple MCP servers in
parallel, leading to incomplete work that never finished. This adds
a mandatory rule: work on exactly one MCP/LSP server end-to-end
(build, handshake, live client invocation with real data, status
update, commit) before starting the next.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-07-30 16:42:57 -05:00
mrcharles d4e9fd693f feat: add discourse-cli for Discourse forum integration
A standalone CLI container (following the redmine-cli pattern) that
lets AI agents and humans read, post, reply, search, and discuss on a
Discourse forum via the REST API. Uses three environment variables
(DISCOURSE_URL, DISCOURSE_API_KEY, DISCOURSE_API_USERNAME) for auth.

Commands: whoami, categories, cat-info, topics/ls, show, create, reply,
update, delete, search, notifications.

Fully validated end-to-end (10/10 checks) against a live Discourse
instance: credential auth, whoami, categories, topics, show, search,
and full write cycle (create + reply + update + delete).

Includes validate.sh for repeatable end-to-end testing using
containerized curl + the CLI image.

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-07-30 16:39:17 -05:00
mrcharles 9d6764be11 fix: add honest status reporting protocol to prevent false validation claims
AI agents were marking services as "validated" or "working" after only
running structural tests (build + --help) without any live interaction
against the real external system. This adds a mandatory section to
AGENTS.md that:

- Distinguishes "built", "structural test", and "validated" as distinct
  states with concrete examples
- Prohibits using "validated"/"working"/"done"/"complete" for credential-
  dependent services until a live operation succeeds
- Requires obtaining credentials and running real read/write operations
  before claiming completion
- Enforces todo-list honesty: items can only be marked completed when
  their acceptance criterion is actually met

💘 Generated with Crush

Assisted-by: Crush:glm-5.2
2026-07-30 16:39:07 -05:00
mrcharles 4ad44bbe9d fix: update Dockerfiles for upstream changes and build fixes
- gitea-mcp, mcp-grafana: bump Go from 1.24 to 1.26 (upstream go.mod
  now requires >= 1.26.x)
- proxmox-mcp: constrain requires-python to >=3.10 for fastmcp compat
- elasticsearch-mcp: add missing vendor repo to CloneVendorRepos.sh
  (elastic/mcp-server-elasticsearch)
- context7-mcp: add .npmrc to disable pnpm v10 strict built-deps check
- drawio-mcp: exclude dev-proxy package from build (upstream TS errors),
  build with pnpm v10 + .npmrc
- discourse-mcp, ghidra-mcp: remain blocked (upstream issues)
2026-07-30 15:47:30 -05:00
mrcharles b595e26d62 fix: add profile flags to all docker compose commands and complete env vars
All services in docker-compose.yml are behind profiles (dev/ops), but the
maintenance scripts called docker compose without --profile, so no services
were detected or built. Added --profile dev --profile ops to BuildAll.sh,
StatusCheck.sh, and all relevant maintenance.sh commands.

Also:
- Added missing env vars for 8 newer servers to .env.example (beszel, gitea,
  grafana, ha, limesurvey, linkwarden, superset)
- Fixed maintenance.sh validate to use validate-all.sh instead of nonexistent
  validate-mcp.sh
2026-07-30 12:15:49 -05:00
mrcharles b3a810ba07 fix: make all scripts resolve paths relative to repo root
CloneVendorRepos.sh and CleanVendor.sh resolved vendor/ relative to scripts/
instead of the repo root, so vendor repos were cloned to the wrong location
(scripts/vendor) and docker-compose builds (which expect ./vendor at repo
root) would fail. BuildAll.sh and StatusCheck.sh had the same issue, cd-ing
to scripts/ instead of the repo root before running docker compose.

Also updated validate-all.sh with the 10 newer MCP servers that were added
after the script was last maintained (firefly-iii, actual, paperless, beszel,
gitea, grafana, ha, limesurvey, linkwarden, superset) plus added missing
env vars for audiobook-mcp and drawio-mcp.
2026-07-30 12:12:43 -05:00
mrcharles 237f345582 fix: correct crush.json wrapper script paths
All wrapper script paths referenced /home/charles/Projects/ (previous
machine/user) instead of the actual /home/reachableceo/projects/. Every
LSP and MCP entry in crush.json was broken. Fixed 39 path references.
2026-07-30 12:12:17 -05:00
48 changed files with 2295 additions and 105 deletions
+31
View File
@@ -98,3 +98,34 @@ PAPERLESS_TOKEN=dummy-paperless-token-replace-with-real
ACTUAL_SERVER_URL=https://actual.example.com
ACTUAL_PASSWORD=dummy-password-replace-with-real
ACTUAL_BUDGET_SYNC_ID=dummy-budget-sync-id
# Beszel
BESZEL_URL=https://beszel.example.com
BESZEL_USERNAME=dummy-username
BESZEL_PASSWORD=dummy-password-replace-with-real
# Gitea
GITEA_URL=https://gitea.example.com
GITEA_TOKEN=dummy-token-replace-with-real
# Grafana
GRAFANA_URL=https://grafana.example.com
GRAFANA_TOKEN=dummy-token-replace-with-real
# Home Assistant
HOMEASSISTANT_URL=https://ha.example.com
HOMEASSISTANT_TOKEN=dummy-token-replace-with-real
# Limesurvey
LIMESURVEY_URL=https://limesurvey.example.com
LIMESURVEY_USERNAME=admin
LIMESURVEY_PASSWORD=dummy-password-replace-with-real
# Linkwarden
LINKWARDEN_URL=https://linkwarden.example.com
LINKWARDEN_TOKEN=dummy-token-replace-with-real
# Superset
SUPERSET_URL=https://superset.example.com
SUPERSET_USERNAME=admin
SUPERSET_PASSWORD=dummy-password-replace-with-real
+48 -2
View File
@@ -51,8 +51,8 @@ ALL work performed on this project MUST be documented in [JOURNAL.md](JOURNAL.md
- **Custom images**: publish to the Gitea container registry at
`git.knownelement.com/reachableceo/<image>:<tag>` (Docker is pre-authenticated).
- **Standalone CLI tools**: containerized CLI tools (not MCP/LSP servers) live
in their own top-level directory (e.g. `redmine-cli/`). Invoked on-demand via
`docker run --env-file .env <image> <command>`, not as long-running services.
under `tooling-cli/` (e.g. `tooling-cli/redmine/`). Invoked on-demand via
`docker run --env-file ~/.creds/<service>.env <image> <command>`, not as long-running services.
## Crush Configuration
@@ -100,6 +100,31 @@ For each agent, verify in order:
**CRITICAL**: Items 1-3 are NON-NEGOTIABLE. Do not mark a server as "working" without protocol validation.
## Honest Status Reporting (MANDATORY)
AI agents MUST distinguish between these distinct states and never blur them:
| State | Meaning | Example |
|-------|---------|---------|
| **Built** | Image/container compiles or assembles without errors | `docker build` exits 0 |
| **Structural test passed** | Binary runs, `--help` works, error paths return correct codes | CLI prints usage, missing-env exits 2 |
| **Validated / Working** | A **live** interaction against a real external system succeeded | `whoami` returns real user data from the live API |
**A structural test is NOT validation.** Building + `--help` does not make a tool "validated" or "working".
### Rules
1. **Never use the words "validated", "working", "done", or "complete"** for any service that depends on external credentials or a live endpoint until you have:
- **Obtained** the real credentials / endpoint (not placeholder/dummy values).
- **Executed** at least one live read or write operation against the real system.
- **Observed** a successful response with real data.
2. **Never claim validation is impossible then state the tool is done.** If you lack credentials, say *"blocked — needs credentials to validate"* and obtain them. Do not imply completion by omission.
3. **Todo list honesty:** A todo item may only be marked `completed` when its acceptance criterion has actually been met. "Build and test" requires a passing test, not a successful build alone.
4. **When blocked on validation:** Stop, state the exact blocker (e.g. "need DISCOURSE_API_KEY to run live whoami test"), and obtain it before proceeding or declaring the task done.
## MCP Handshake Command
```bash
@@ -134,6 +159,27 @@ git config core.hooksPath .githooks
If you cannot complete an item (e.g., no network access for push), explicitly state which items are blocked and why.
## One Server At A Time (MANDATORY)
AI agents MUST work on exactly **one MCP/LSP server at a time, end-to-end** before
starting the next one. This includes:
1. **Build** the container
2. **Run** the container and validate the protocol handshake
3. **Spin up a live Crush or bash session** and actually invoke the server through
the real client pipeline
4. **Observe a successful real tool call** (not just a handshake — an actual tool
invocation that returns real data or completes a real action)
5. **Update STATUS.md and JOURNAL.md** for that server
6. **Commit and push** before moving on
**NEVER** attempt to build/validate multiple MCP servers in parallel or batch them.
**NEVER** skip from "built" to the next server without completing steps 3-4 (real
client invocation with real data).
The pattern of "build all, then validate all" leads to incomplete work that never
finishes. One server, fully done, then the next.
## Related Documentation
- [README.md](README.md) - Project overview, server inventory, installation, usage
+64
View File
@@ -1924,3 +1924,67 @@ is the canonical home for all AI middleware going forward.
`tea` commands run on the host; everything else is containerized. Custom images
are published to the Gitea container registry at
`git.knownelement.com/reachableceo/<image>`.
## 2026-07-30
### [ADR] discourse-cli: Standalone CLI for Discourse (following redmine-cli pattern)
**Date/Time**: 2026-07-30 5:50 PM EST
**Type**: Implementation
**Status**: Accepted
**Context**: The existing discourse-mcp was BLOCKED (TS2345 build error), but
the redmine-cli pattern (standalone CLI container, on-demand via docker run)
provides a reliable alternative for Crush (which has Docker/CLI access).
**Decision**: Built discourse-cli/ as a standalone Python CLI with full read/write
coverage: whoami, categories, topics, show, create, reply, update, delete, search,
notifications. Uses requests library with JSON body encoding for POST/PUT/DELETE.
Validated 10/10 against community.turnsys.com.
**Key insight**: Python `requests` with `data={nested}` stringifies dicts instead
of form-encoding them. Must use `json=data` for nested payloads to Discourse API.
The `/session/current.json` endpoint doesn't work with API keys (returns 403);
use `/notifications.json` or `/u/<username>.json` instead for connection tests.
### [ADR] discourse-mcp: Fixed TS2345 build error
**Date/Time**: 2026-07-30 5:50 PM EST
**Type**: Bug Fix
**Status**: Accepted
**Context**: discourse-mcp was BLOCKED with TypeScript TS2345 error in
src/tools/remote/tool_exec_api.ts:58. The Dockerfile was forcing
`@modelcontextprotocol/sdk@^1.23.0` via `pnpm add`, but the upstream code
is only compatible with the lockfile's SDK v1.17.3.
**Decision**: Removed the `pnpm add @modelcontextprotocol/sdk@^1.23.0` line
from the Dockerfile. Using `pnpm install --frozen-lockfile` instead.
**Additional fix**: The discourse-mcp server does NOT read environment variables
directly. It requires CLI args (`--auth-pairs`, `--site`) or a `--profile` JSON
file. Added a docker-entrypoint.sh that converts DISCOURSE_URL/API_KEY/API_USERNAME
env vars into a profile JSON file with auth_pairs and site tethering.
**Key insight**: The upstream server's `buildAuth()` returns `{type: "none"}`
by default. Auth comes from auth_pairs overrides in the profile. The `select_site`
tool is needed when not tethered; when `--site` is provided, the server preselects
the site and hides the select_site tool.
Validated with MCP handshake + live tool calls (search, filter_topics) against
community.turnsys.com.
### Protocol Fix: Honest status reporting + one-server-at-a-time
**Date/Time**: 2026-07-30 5:50 PM EST
**Type**: Process Improvement
**Status**: Accepted
**Context**: AI agents were marking services as "validated" after only running
structural tests (build + --help) without live credentials. Also attempting to
validate multiple MCP servers in parallel, never finishing any.
**Decision**: Added two mandatory rules to AGENTS.md:
1. Honest status reporting: distinguish "built" vs "structural test" vs "validated"
2. One server at a time: build → handshake → live client invocation → status → commit
before starting the next server.
+1 -1
View File
@@ -83,7 +83,7 @@ Legend: ✅ Production Ready | ⚠️ Config Required | ❌ Not Production Ready
#### Communication & Collaboration (3 servers)
| Service | Status | Container Name | Description |
|---------|--------|---------------|-------------|
| discourse-mcp | | kneldevstack-aimiddleware-discourse-mcp | Forum (upstream build error) |
| discourse-mcp | | kneldevstack-aimiddleware-discourse-mcp | Forum (fixed: SDK version + env-var auth) |
| imap-mcp | ❌ | kneldevstack-aimiddleware-imap-mcp | Email (crashes without live IMAP) |
| postizz-mcp | ❌ | kneldevstack-aimiddleware-postizz-mcp | Social media (HTTP transport, not stdio) |
+4 -4
View File
@@ -100,7 +100,7 @@ All 32 vendor repositories have been verified and correctly cloned. CloneVendorR
- ✓ ghidra-mcp: Working (v1.9.4-headless) - 91 REST endpoints for binary analysis - NOT MCP protocol (uses HTTP REST)
**MCP Servers with Build/Runtime Issues:**
- discourse-mcp: BLOCKED - TypeScript TS2345 error (upstream SDK incompatibility, cannot fix)
- discourse-mcp: Working (@discourse/mcp v0.2.9) - requires DISCOURSE_URL, DISCOURSE_API_KEY, DISCOURSE_API_USERNAME env vars
- ✗ reverse-engineering-assistant: BLOCKED - requires specific gradle version matching Ghidra (complex build)
- ✗ penpot-mcp: Build OK, transport mismatch - uses HTTP/WebSocket transport instead of stdio
@@ -122,7 +122,7 @@ All 32 vendor repositories have been verified and correctly cloned. CloneVendorR
| blender-mcp | Built | Container built from source (321MB). Python-based with uv package manager. MCP stdio-based, requires Blender running with addon. Version 1.25.0. |
| cloudron-mcp | Built | Container built from source (374MB). MCP stdio-based, requires CLOUDRON_URL env var. Version 0.1.0. |
| context7-mcp | Built | Container built from source (224MB). Multi-stage Node.js build with tsc. MCP stdio-based, requires UPSTASH_REDIS_REST_URL and TOKEN env vars. Version 2.1.0. |
| discourse-mcp | Build Failed | TypeScript TS2345 error in upstream repository. Cannot build locally. |
| discourse-mcp | Built | Container built from TypeScript source. Fixed: removed SDK override (pnpm add ^1.23.0) that caused TS2345, added entrypoint to convert env vars to profile JSON. MCP stdio-based, auto-tethers to DISCOURSE_URL. Version 0.2.9. Validated with live tool calls against community.turnsys.com. |
| docker-mcp | Built | Container built from Python source (188MB). Uses uv package manager. MCP stdio-based, requires Docker socket mount (/var/run/docker.sock). Version 0.1.0. |
| docspace-mcp | Built | Container built from official ONLYOFFICE TypeScript source (236MB). Uses pnpm package manager. MCP stdio-based, requires DOCSPACE_HOST and DOCSPACE_TOKEN env vars. Version 3.1.0. |
| drawio-mcp | Built | Container built from TypeScript source (302MB). Uses pnpm package manager. MCP stdio-based, requires DRAWIO_URL env var. Version 1.4.0. |
@@ -170,7 +170,7 @@ Host-only - requires KiCAD installed on the host machine. The pcbnew Python modu
The WordPress MCP Adapter is a PHP plugin that must be installed within a running WordPress instance. It requires WordPress + Abilities API plugin + MCP Adapter plugin. Not suitable for containerized standalone deployment.
### discourse-mcp
BLOCKED - Build failed with TypeScript TS2345 error in upstream repository (src/tools/remote/tool_exec_api.ts:58). The code is incompatible with MCP SDK types. Cannot fix locally without upstream changes.
FIXED - Build was failing due to Dockerfile forcing `@modelcontextprotocol/sdk@^1.23.0` via `pnpm add`, which is incompatible with upstream code (locks SDK to v1.17.3). Removed the `pnpm add` override, using `pnpm install --frozen-lockfile` instead. Also added a docker-entrypoint.sh that converts env vars (DISCOURSE_URL, DISCOURSE_API_KEY, DISCOURSE_API_USERNAME) into a profile JSON file with auth_pairs and site tethering. The server does not read env vars directly — it requires CLI args or a profile file. Validated with MCP handshake + live tool calls (search, filter_topics) against community.turnsys.com.
### proxmox-mcp
Fixed by patching pyproject.toml to use 'fastmcp' from PyPI instead of 'mcp @ git+...' and updating imports from 'mcp.server.fastmcp' to 'fastmcp'. Container now builds and runs - requires PROXMOX_MCP_CONFIG env var pointing to a valid config file.
@@ -217,4 +217,4 @@ Standalone containerized CLI tools (not MCP/LSP servers). Invoked on-demand via
| Tool | Image | Status | Notes |
|------|-------|--------|-------|
| redmine-cli | kneldevstack-aimiddleware-redmine-cli | Working | Custom CLI (python-redmine) to list/show/create/update/close Redmine issues. Verified end-to-end against projects.knownelement.com. Requires REDMINE_URL and REDMINE_API_KEY. Source in `redmine-cli/`. |
| redmine-cli | git.knownelement.com/reachableceo/redmine-cli:latest | Working | Custom CLI (python-redmine) to list/show/create/update/close Redmine issues. Verified end-to-end against projects.knownelement.com. Requires REDMINE_URL and REDMINE_API_KEY. Source in `tooling-cli/redmine/`. |
+39 -39
View File
@@ -2,192 +2,192 @@
"$schema": "https://charm.land/crush.json",
"lsp": {
"bash": {
"command": "/home/charles/Projects/KNEL-AIMiddleware/lsp-bash-wrapper.sh"
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/lsp-bash-wrapper.sh"
},
"docker": {
"command": "/home/charles/Projects/KNEL-AIMiddleware/lsp-docker-wrapper.sh"
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/lsp-docker-wrapper.sh"
},
"markdown": {
"command": "/home/charles/Projects/KNEL-AIMiddleware/lsp-marksman-wrapper.sh"
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/lsp-marksman-wrapper.sh"
},
"terraform": {
"command": "/home/charles/Projects/KNEL-AIMiddleware/lsp-terraform-wrapper.sh"
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/lsp-terraform-wrapper.sh"
}
},
"mcp": {
"audiobook": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-audiobook-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-audiobook-wrapper.sh",
"timeout": 60
},
"bitwarden": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-bitwarden-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-bitwarden-wrapper.sh",
"timeout": 60
},
"blender": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-blender-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-blender-wrapper.sh",
"timeout": 180
},
"cloudron": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-cloudron-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-cloudron-wrapper.sh",
"timeout": 60
},
"context7": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-context7-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-context7-wrapper.sh",
"timeout": 60
},
"docker": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-docker-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-docker-wrapper.sh",
"timeout": 180
},
"drawio": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-drawio-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-drawio-wrapper.sh",
"timeout": 60
},
"elasticsearch": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-elasticsearch-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-elasticsearch-wrapper.sh",
"timeout": 60
},
"freecad": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-freecad-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-freecad-wrapper.sh",
"timeout": 180
},
"ghost": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-ghost-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-ghost-wrapper.sh",
"timeout": 60
},
"gimp": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-gimp-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-gimp-wrapper.sh",
"timeout": 180
},
"imap": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-imap-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-imap-wrapper.sh",
"timeout": 60
},
"kubernetes": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-kubernetes-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-kubernetes-wrapper.sh",
"timeout": 60
},
"docspace": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-docspace-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-docspace-wrapper.sh",
"timeout": 60
},
"webserial": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-webserial-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-webserial-wrapper.sh",
"timeout": 60
},
"proxmox": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-proxmox-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-proxmox-wrapper.sh",
"timeout": 60
},
"terraform": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-terraform-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-terraform-wrapper.sh",
"timeout": 60
},
"nextcloud": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-nextcloud-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-nextcloud-wrapper.sh",
"timeout": 60
},
"matomo": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-matomo-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-matomo-wrapper.sh",
"timeout": 60
},
"snipeit": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-snipeit-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-snipeit-wrapper.sh",
"timeout": 60
},
"redmine": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-redmine-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-redmine-wrapper.sh",
"timeout": 60
},
"ansible": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-ansible-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-ansible-wrapper.sh",
"timeout": 60
},
"actual": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-actual-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-actual-wrapper.sh",
"timeout": 60
},
"beszel": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-beszel-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-beszel-wrapper.sh",
"timeout": 60
},
"gitea": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-gitea-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-gitea-wrapper.sh",
"timeout": 60
},
"grafana": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-grafana-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-grafana-wrapper.sh",
"timeout": 60
},
"homeassistant": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-ha-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-ha-wrapper.sh",
"timeout": 60
},
"limesurvey": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-limesurvey-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-limesurvey-wrapper.sh",
"timeout": 60
},
"linkwarden": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-linkwarden-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-linkwarden-wrapper.sh",
"timeout": 60
},
"superset": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-superset-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-superset-wrapper.sh",
"timeout": 60
},
"postizz": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-postizz-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-postizz-wrapper.sh",
"timeout": 60
},
"discourse": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-discourse-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-discourse-wrapper.sh",
"timeout": 60
},
"penpot": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-penpot-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-penpot-wrapper.sh",
"timeout": 60
},
"firefly-iii": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-firefly-iii-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-firefly-iii-wrapper.sh",
"timeout": 60
},
"paperless": {
"type": "stdio",
"command": "/home/charles/Projects/KNEL-AIMiddleware/mcp-paperless-wrapper.sh",
"command": "/home/reachableceo/projects/KNEL-AIMiddleware/mcp-paperless-wrapper.sh",
"timeout": 60
}
}
+6 -9
View File
@@ -282,10 +282,8 @@ services:
dockerfile: ../../dockerfiles/discourse-mcp/Dockerfile
container_name: kneldevstack-aimiddleware-discourse-mcp
restart: unless-stopped
environment:
- DISCOURSE_URL=${DISCOURSE_URL}
- DISCOURSE_API_KEY=${DISCOURSE_API_KEY}
- DISCOURSE_API_USERNAME=${DISCOURSE_API_USERNAME}
env_file:
- /home/reachableceo/.creds/discourse.env
profiles:
- ops
@@ -398,10 +396,10 @@ services:
dockerfile: ../../dockerfiles/mcp-redmine/Dockerfile
container_name: kneldevstack-aimiddleware-mcp-redmine
restart: "no"
env_file:
- /home/reachableceo/.creds/redmine.env
environment:
- PYTHONUNBUFFERED=1
- REDMINE_URL=${REDMINE_URL}
- REDMINE_API_KEY=${REDMINE_API_KEY}
profiles:
- ops
@@ -609,11 +607,10 @@ services:
dockerfile: ../../dockerfiles/beszel-mcp/Dockerfile
container_name: kneldevstack-aimiddleware-beszel-mcp
restart: "no"
env_file:
- /home/reachableceo/.creds/beszel.env
environment:
- PYTHONUNBUFFERED=1
- BESZEL_URL=${BESZEL_URL}
- BESZEL_USERNAME=${BESZEL_USERNAME}
- BESZEL_PASSWORD=${BESZEL_PASSWORD}
profiles:
- ops
+4 -2
View File
@@ -1,13 +1,15 @@
FROM node:22-alpine
# Install pnpm for monorepo
# Install pnpm
RUN npm install -g pnpm
WORKDIR /app
COPY package.json pnpm-lock.yaml ./
RUN pnpm install --frozen-lockfile
# Disable strict built-dependencies check (pnpm v10 exits on ignored builds)
RUN echo "strict-built-dependencies=false" > .npmrc && \
pnpm install --frozen-lockfile
COPY . .
+25 -3
View File
@@ -7,12 +7,34 @@ RUN corepack enable && corepack prepare pnpm@10.14.0 --activate
COPY package.json pnpm-lock.yaml ./
# Install dependencies with pinned SDK version
RUN pnpm add @modelcontextprotocol/sdk@^1.23.0
# Install dependencies from lockfile (pins SDK to upstream-compatible version)
RUN pnpm install --frozen-lockfile
COPY . .
RUN pnpm build
CMD ["node", "dist/index.js"]
# Entrypoint script converts env vars to CLI args (profile JSON file)
RUN cat <<'ENTRYEOF' > /docker-entrypoint.sh
#!/bin/sh
set -e
if [ -n "${DISCOURSE_URL}" ] && [ -n "${DISCOURSE_API_KEY}" ]; then
echo -n '{"auth_pairs":[{"site":"' > /tmp/profile.json
echo -n "${DISCOURSE_URL}" >> /tmp/profile.json
echo -n '","api_key":"' >> /tmp/profile.json
echo -n "${DISCOURSE_API_KEY}" >> /tmp/profile.json
echo -n '"' >> /tmp/profile.json
if [ -n "${DISCOURSE_API_USERNAME}" ]; then
echo -n ',"api_username":"' >> /tmp/profile.json
echo -n "${DISCOURSE_API_USERNAME}" >> /tmp/profile.json
echo -n '"' >> /tmp/profile.json
fi
echo -n '}]}' >> /tmp/profile.json
exec node dist/index.js --profile /tmp/profile.json --site "${DISCOURSE_URL}" --allow-writes --read-only false "$@"
else
exec node dist/index.js "$@"
fi
ENTRYEOF
RUN chmod +x /docker-entrypoint.sh
ENTRYPOINT ["/docker-entrypoint.sh"]
@@ -0,0 +1,32 @@
#!/bin/sh
# Convert env vars to discourse-mcp CLI arguments, then exec the server.
#
# Env vars:
# DISCOURSE_URL - base URL of the Discourse site
# DISCOURSE_API_KEY - API key
# DISCOURSE_API_USERNAME - username for the API key
#
# When DISCOURSE_URL is set, the server is tethered to that site with auth
# pre-configured. Otherwise, the client must call discourse_select_site.
set -e
ARGS=""
if [ -n "${DISCOURSE_URL}" ]; then
ARGS="${ARGS} --site ${DISCOURSE_URL}"
if [ -n "${DISCOURSE_API_KEY}" ]; then
AUTH_PAIR="{\"site\":\"${DISCOURSE_URL}\""
AUTH_PAIR="${AUTH_PAIR},\"api_key\":\"${DISCOURSE_API_KEY}\""
if [ -n "${DISCOURSE_API_USERNAME}" ]; then
AUTH_PAIR="${AUTH_PAIR},\"api_username\":\"${DISCOURSE_API_USERNAME}\""
fi
AUTH_PAIR="${AUTH_PAIR}}"
ARGS="${ARGS} --auth-pairs '[${AUTH_PAIR}]'"
fi
fi
ARGS="${ARGS} --allow-writes --read-only false"
exec eval node dist/index.js ${ARGS} "$@"
+7 -3
View File
@@ -7,10 +7,14 @@ WORKDIR /app
COPY package.json pnpm-lock.yaml ./
RUN pnpm install --frozen-lockfile
# Disable strict built-dependencies check (pnpm v10 exits on ignored builds)
RUN echo "strict-built-dependencies=false" > .npmrc && \
pnpm install --no-frozen-lockfile
COPY . .
RUN pnpm run build
# Build all packages except dev-proxy (has upstream TS errors).
# Plugin must be built before server (server copies plugin output).
RUN pnpm -r --filter '!drawio-mcp-dev-proxy' build
CMD ["node", "build/index.js"]
CMD ["node", "packages/drawio-mcp-server/build/index.js"]
+1 -1
View File
@@ -1,4 +1,4 @@
FROM golang:1.24-alpine AS builder
FROM golang:1.26-alpine AS builder
WORKDIR /app
+1 -1
View File
@@ -1,4 +1,4 @@
FROM golang:1.24-alpine AS builder
FROM golang:1.26-alpine AS builder
WORKDIR /app
+3
View File
@@ -12,6 +12,9 @@ COPY . .
# Replace mcp git dependency with fastmcp from PyPI
RUN sed -i 's|"mcp @ git+https://github.com/modelcontextprotocol/python-sdk.git"|"fastmcp"|' pyproject.toml
# Fix requires-python for fastmcp compatibility (needs >=3.10)
RUN sed -i 's/requires-python = ">=3.9"/requires-python = ">=3.10"/' pyproject.toml
# Patch imports to use fastmcp instead of mcp.server.fastmcp
RUN find . -name "*.py" -exec sed -i 's/from mcp\.server\.fastmcp/from fastmcp/g' {} \;
+6 -6
View File
@@ -65,7 +65,7 @@ build_service() {
exit 1
fi
echo -e "${GREEN}=== Building $service ===${NC}"
docker compose build "$service"
docker compose --profile dev --profile ops build "$service"
}
# Function to clean vendor directory
@@ -89,19 +89,19 @@ status() {
# Function to validate MCP servers
validate() {
echo -e "${GREEN}=== Validating MCP Servers ===${NC}"
./scripts/validate-mcp.sh
./scripts/validate-all.sh mcp
}
# Function to show logs
logs() {
echo -e "${GREEN}=== Service Logs ===${NC}"
docker compose logs -f
docker compose --profile dev --profile ops logs -f
}
# Function to show process status
ps() {
echo -e "${GREEN}=== Service Process Status ===${NC}"
docker compose ps
docker compose --profile dev --profile ops ps
}
# Function to start services
@@ -126,8 +126,8 @@ rebuild() {
exit 1
fi
echo -e "${GREEN}=== Rebuilding $service ===${NC}"
docker compose build --no-cache "$service"
docker compose up -d "$service"
docker compose --profile dev --profile ops build --no-cache "$service"
docker compose --profile dev --profile ops up -d "$service"
}
# Main script logic
+8
View File
@@ -14,6 +14,14 @@ if docker ps -a --filter "name=${CONTAINER_NAME}" --format '{{.Names}}' | grep -
done
fi
# Load credentials from central store
CREDS_FILE="/home/reachableceo/.creds/beszel.env"
if [ -f "$CREDS_FILE" ]; then
set -a
. "$CREDS_FILE"
set +a
fi
# Start MCP server with explicit name and environment
exec docker run -i --rm --name "${CONTAINER_NAME}" \
-e BESZEL_URL="${BESZEL_URL}" \
+8
View File
@@ -14,6 +14,14 @@ if docker ps -a --filter "name=${CONTAINER_NAME}" --format '{{.Names}}' 2>/dev/n
done
fi
# Load credentials from central store
CREDS_FILE="/home/reachableceo/.creds/discourse.env"
if [ -f "$CREDS_FILE" ]; then
set -a
. "$CREDS_FILE"
set +a
fi
# Start MCP server with explicit name and environment variables
exec docker run -i --rm --name "${CONTAINER_NAME}" \
-e "DISCOURSE_URL=${DISCOURSE_URL:-}" \
+8
View File
@@ -14,6 +14,14 @@ if docker ps -a --filter "name=${CONTAINER_NAME}" --format '{{.Names}}' 2>/dev/n
done
fi
# Load credentials from central store
CREDS_FILE="/home/reachableceo/.creds/redmine.env"
if [ -f "$CREDS_FILE" ]; then
set -a
. "$CREDS_FILE"
set +a
fi
# Start MCP server with explicit name and environment variables
exec docker run -i --rm --name "${CONTAINER_NAME}" \
-e "REDMINE_URL=${REDMINE_URL:-}" \
+5 -5
View File
@@ -17,16 +17,16 @@ echo ""
echo "Building all services..."
echo ""
# Change to project directory
cd "$(dirname "${BASH_SOURCE[0]}")"
# Change to project root (parent of scripts/)
cd "$(dirname "${BASH_SOURCE[0]}")/.."
# Get list of all services from docker-compose.yml
SERVICES=$(docker compose config --services)
# Get list of all services from docker-compose.yml (include all profiles)
SERVICES=$(docker compose --profile dev --profile ops config --services | sort)
# Build each service
for SERVICE in $SERVICES; do
echo -e "${BLUE}Building: ${SERVICE}${NC}"
if docker compose build "$SERVICE" 2>&1 | tail -5; then
if docker compose --profile dev --profile ops build "$SERVICE" 2>&1 | tail -5; then
echo -e "${GREEN}${NC} Successfully built $SERVICE"
else
echo -e "${RED}${NC} Failed to build $SERVICE"
+3 -2
View File
@@ -22,8 +22,9 @@ if [ "$CONFIRM" != "yes" ]; then
exit 0
fi
# Vendor directory
VENDOR_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/vendor"
# Vendor directory (repo root is parent of scripts/)
REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
VENDOR_DIR="$REPO_ROOT/vendor"
echo ""
echo "Removing vendor directories..."
+4 -2
View File
@@ -11,8 +11,9 @@ GREEN='\033[0;32m'
YELLOW='\033[1;33m'
NC='\033[0m' # No Color
# Vendor directory
VENDOR_DIR="$(cd "$(dirname "${BASH_SOURCE[0]}")" && pwd)/vendor"
# Vendor directory (repo root is parent of scripts/)
REPO_ROOT="$(cd "$(dirname "${BASH_SOURCE[0]}")/.." && pwd)"
VENDOR_DIR="$REPO_ROOT/vendor"
echo -e "${GREEN}=== KNEL-AIMiddleware Vendor Repository Clone Script ===${NC}"
echo ""
@@ -90,6 +91,7 @@ clone_repo "gitea-mcp" "https://gitea.com/gitea/gitea-mcp.git"
clone_repo "limesurvey-mcp" "https://github.com/TonisOrmisson/limesurvey-mcp.git"
clone_repo "linkwarden-mcp-server" "https://github.com/irfansofyana/linkwarden-mcp-server.git"
clone_repo "firefly-iii-mcp" "https://github.com/etnperlong/firefly-iii-mcp.git"
clone_repo "mcp-server-elasticsearch" "https://github.com/elastic/mcp-server-elasticsearch.git"
echo -e "${GREEN}=== All repositories cloned successfully! ===${NC}"
echo ""
+4 -4
View File
@@ -15,11 +15,11 @@ NC='\033[0m' # No Color
echo -e "${BLUE}=== KNEL-AIMiddleware Service Status Check ===${NC}"
echo ""
# Change to project directory
cd "$(dirname "${BASH_SOURCE[0]}")"
# Change to project root (parent of scripts/)
cd "$(dirname "${BASH_SOURCE[0]}")/.."
# Get list of all services from docker-compose.yml
SERVICES=$(docker compose config --services | sort)
# Get list of all services from docker-compose.yml (include all profiles)
SERVICES=$(docker compose --profile dev --profile ops config --services | sort)
# Counters
TOTAL=0
+16 -2
View File
@@ -72,12 +72,26 @@ MCP_SERVERS=(
["kneldevstack-aimiddleware-bitwarden-mcp"]="15 BITWARDEN_CLIENT_ID=dummy BITWARDEN_CLIENT_SECRET=dummy BITWARDEN_PASSWORD=dummy BITWARDEN_SERVER_URL=https://vault.bitwarden.com"
# Productivity & Automation
["kneldevstack-aimiddleware-audiobook-mcp"]="10"
["kneldevstack-aimiddleware-audiobook-mcp"]="10 AUDIOBOOK_ROOT=/tmp"
["kneldevstack-aimiddleware-snipeit-mcp"]="10 SNIPEIT_URL=https://snipeit.example.com SNIPEIT_TOKEN=dummy"
["kneldevstack-aimiddleware-mcp-redmine"]="10 REDMINE_URL=https://redmine.example.com REDMINE_API_KEY=dummy"
["kneldevstack-aimiddleware-mcp-ansible"]="10"
["kneldevstack-aimiddleware-elasticsearch-mcp"]="10 ELASTICSEARCH_URL=http://localhost:9200 ELASTICSEARCH_USERNAME=dummy ELASTICSEARCH_PASSWORD=dummy"
["kneldevstack-aimiddleware-drawio-mcp"]="10"
["kneldevstack-aimiddleware-drawio-mcp"]="10 DRAWIO_URL=https://drawio.example.com"
# Financial & Budgeting
["kneldevstack-aimiddleware-firefly-iii-mcp"]="10 FIREFLY_III_BASE_URL=https://firefly.example.com FIREFLY_III_PAT=dummy"
["kneldevstack-aimiddleware-actual-mcp"]="10 ACTUAL_SERVER_URL=https://actual.example.com ACTUAL_PASSWORD=dummy ACTUAL_BUDGET_SYNC_ID=dummy"
["kneldevstack-aimiddleware-paperless-mcp"]="10"
# Additional Productivity
["kneldevstack-aimiddleware-beszel-mcp"]="10 BESZEL_URL=https://beszel.example.com BESZEL_USERNAME=dummy BESZEL_PASSWORD=dummy"
["kneldevstack-aimiddleware-gitea-mcp"]="10 GITEA_URL=https://gitea.example.com GITEA_TOKEN=dummy"
["kneldevstack-aimiddleware-mcp-grafana"]="10 GRAFANA_URL=https://grafana.example.com GRAFANA_TOKEN=dummy"
["kneldevstack-aimiddleware-ha-mcp"]="10 HOMEASSISTANT_URL=https://ha.example.com HOMEASSISTANT_TOKEN=dummy"
["kneldevstack-aimiddleware-limesurvey-mcp"]="10 LIMESURVEY_URL=https://limesurvey.example.com LIMESURVEY_USERNAME=dummy LIMESURVEY_PASSWORD=dummy"
["kneldevstack-aimiddleware-linkwarden-mcp"]="10 LINKWARDEN_URL=https://linkwarden.example.com LINKWARDEN_TOKEN=dummy"
["kneldevstack-aimiddleware-superset-mcp"]="10 SUPERSET_URL=https://superset.example.com SUPERSET_USERNAME=dummy SUPERSET_PASSWORD=dummy"
# Additional Tools
["kneldevstack-aimiddleware-penpot-mcp"]="10 PENPOT_URL=https://design.penpot.app PENPOT_TOKEN=dummy"
+6
View File
@@ -0,0 +1,6 @@
FROM node:22-slim
ARG BW_CLI_VERSION=2026.7.0
RUN npm install -g @bitwarden/cli@${BW_CLI_VERSION} && npm cache clean --force
ENTRYPOINT ["bw"]
+87
View File
@@ -0,0 +1,87 @@
# KNELCredsManager
Centralized credential management for KNEL infrastructure. Stores service
credentials on disk in `~/.creds/` and provides a containerized Bitwarden CLI
for future migration to a password manager.
## Credential store layout
All credentials live in `~/.creds/` as flat `.env` files, one per service:
```
~/.creds/
├── beszel.env # Beszel monitoring (hub URL set, auth TBD — see #406)
├── discourse.env # Discourse forum API keys + admin key
├── phpipam.env # phpIPAM app_id + app_code
├── redmine.env # Redmine REST API key
├── technitium.env # Technitium DNS API key + admin password
└── uptime-kuma.env # Uptime Kuma push/API key
```
Permissions: directory `700`, files `600` (owner read/write only).
### Consumers
All credential consumers source from `~/.creds/`:
| Service | Wrapper / consumer | Mechanism |
|---|---|---|
| Redmine | `redmine-cli` container (`tooling-cli/redmine/`) | `docker run --env-file ~/.creds/redmine.env` |
| Redmine MCP | `mcp-redmine-wrapper.sh` | `set -a; . ~/.creds/redmine.env; set +a` |
| Discourse | `discourse-cli` container (`tooling-cli/discourse/`) | `docker run --env-file ~/.creds/discourse.env` |
| Discourse MCP | `mcp-discourse-wrapper.sh` | `set -a; . ~/.creds/discourse.env; set +a` |
| Beszel MCP | `mcp-beszel-wrapper.sh` | `set -a; . ~/.creds/beszel.env; set +a` |
| Uptime Kuma | (no active consumer yet) | Direct env reference |
| Technitium | (no active consumer yet) | Direct env reference |
| phpIPAM | (no active consumer yet) | Direct env reference |
## Bitwarden CLI
The Bitwarden CLI runs in a pinned Docker container — no host installation
required. Built from this directory's Dockerfile.
### Build
```bash
docker build -t reachableceo-bw-cli:2026.7.0 .
```
### Wrapper
```bash
# Symlink the wrapper onto PATH
ln -sf ~/projects/KNEL-AIMiddleware/tooling-cli/KNELCredsManager/scripts/bw ~/.local/bin/bw
```
### Usage
```bash
bw login # interactive first-time login
export BW_SESSION=$(bw unlock --raw) # unlock and capture session
bw sync # sync vault
bw list items # list all vault items
bw get item <name> # get a specific item
bw status # check auth/session status
```
Session state persists in `~/.local/share/bw-cli/` across container runs.
### MCP integration (machine-to-machine)
For agent automation without interactive login, the KNEL-AIMiddleware fleet
includes `mcp-bitwarden-wrapper.sh` which uses `BITWARDEN_CLIENT_ID` /
`BITWARDEN_CLIENT_SECRET` / `BITWARDEN_PASSWORD` env vars (machine account
auth). That is a separate integration from this CLI wrapper.
## Build arguments
| Arg | Default | Description |
|---|---|---|
| `BW_CLI_VERSION` | `2026.7.0` | Pinned @bitwarden/cli npm version |
## Environment variables
| Variable | Default | Description |
|---|---|---|
| `BW_CLI_IMAGE` | `reachableceo-bw-cli:2026.7.0` | Override image tag |
| `BW_SESSION` | (unset) | Session key from `bw unlock` |
+48
View File
@@ -0,0 +1,48 @@
#!/usr/bin/env bash
# bw — Bitwarden CLI wrapper (Docker containerized, host stays clean).
#
# Usage:
# bw login (interactive — prompts for email/master password/2FA)
# bw unlock (interactive — prints export BW_SESSION=... line)
# bw unlock --raw (prints only the session key, for scripting)
# bw list items
# bw get item <name-or-id>
# bw sync
# bw status
#
# Session management:
# After `bw unlock`, capture the session key:
# export BW_SESSION=$(bw unlock --raw)
# The wrapper passes BW_SESSION through automatically if already set.
#
# Data persistence:
# BW data.json lives at ~/.local/share/bw-cli/ mounted into the container,
# so login state persists across invocations.
set -euo pipefail
IMAGE="${BW_CLI_IMAGE:-reachableceo-bw-cli:2026.7.0}"
DATA_DIR="${HOME}/.local/share/bw-cli"
mkdir -p "$DATA_DIR"
# Detect TTY for interactive commands (login, unlock)
INTERACTIVE=""
if [ -t 0 ] && [ -t 1 ]; then
INTERACTIVE="-it"
fi
# Pass BW_SESSION through if set
SESSION_ARGS=()
if [ -n "${BW_SESSION:-}" ]; then
SESSION_ARGS+=( -e "BW_SESSION=${BW_SESSION}" )
fi
CONTAINER_NAME="reachableceo-bw-cli-$(date +%s)"
exec docker run --rm $INTERACTIVE \
--user "$(id -u):$(id -g)" \
--name "$CONTAINER_NAME" \
-e HOME=/home/bw \
-v "${DATA_DIR}:/home/bw/.config" \
"${SESSION_ARGS[@]}" \
"$IMAGE" "$@"
+5
View File
@@ -0,0 +1,5 @@
# Discourse connection details
# Copy this file to .env and fill in real values. The .env file is gitignored.
DISCOURSE_URL=https://discourse.example.com
DISCOURSE_API_KEY=your-api-key-here
DISCOURSE_API_USERNAME=your-username
+8
View File
@@ -0,0 +1,8 @@
# Secrets - never commit
.env
# OS / editor cruft
.DS_Store
*.swp
*.swo
*~
+346
View File
@@ -0,0 +1,346 @@
# AGENTS.md — Discourse CLI tooling
This directory holds the `discourse-cli` Docker container — a thin Python CLI
over the Discourse REST API. Every Crush session that works on or with this
tool should read this file first: how to invoke it, what commands exist, what
gotchas to avoid, and the patterns to follow for common forum operations.
## Connection & invocation
Invoke the real container with `docker run`. Credentials come from the
centralized credential store (see `tooling-cli/KNELCredsManager`):
```bash
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest <command>
```
There is intentionally **no `bin/` wrapper** — invoke the real container, as
the project-wide rules require. For readability in an interactive session you
may define a throwaway shell alias (do not commit one):
```bash
alias discourse='docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest'
```
The examples below write `discourse <cmd>` for brevity; expand it to the full
`docker run` line (or set the alias) before running.
| Item | Value |
|------|-------|
| Instance | `https://community.turnsys.com` |
| API user | Charles N Wyble (id 2, username `reachableceo`, trust level 4) |
| Admin? | **No** — trust 4 (Leader) but not staff/admin. Admin-only endpoints will 403. |
| Credentials | `~/.creds/discourse.env` (`DISCOURSE_URL`, `DISCOURSE_API_KEY`, `DISCOURSE_API_USERNAME`) |
| CLI image | `git.knownelement.com/reachableceo/discourse-cli:latest` |
| Source | `src/discourse_cli.py` in this directory |
To use a locally built image instead of the registry image, either set
`DISCOURSE_CLI_IMAGE` or swap the tag for `kneldevstack-aimiddleware-discourse-cli:latest`.
## Quick start
```bash
# Connection sanity check (run this first in any session):
discourse whoami
# List categories:
discourse categories
# Latest topics (or in a category):
discourse ls
discourse ls -c general
# Look at a topic:
discourse show 42
```
## Command reference
| Command | Description |
|---------|-------------|
| `whoami` | Authenticated user + connection test. |
| `categories` | List categories (id, slug, topic count, name). |
| `cat-info <id\|slug>` | Show details of a single category. |
| `topics` (`ls`) | List topics (latest, or filtered). Filters below. |
| `show <topic_id>` | Full topic detail incl. all posts (HTML stripped to text). |
| `create` | Create a new topic (first post). |
| `reply <topic_id>` | Reply to a topic. |
| `update <post_id>` | Edit an existing post's body. |
| `delete <post_id>` | Delete a post. |
| `search <query>` | Search the forum. |
| `notifications` | List your notifications. |
### `topics` / `ls` filters
- `-c, --category <id|slug>` — only topics in a category
- `-n, --new` — only new (to you) topics
- `-u, --unread` — only unread topics
- `-p, --page <n>` — page number (default 0)
### `create` options
- `-t, --title` **required**
- `-b, --body` **required** (raw markdown/text — this becomes the first post)
- `-c, --category <id|slug>`
- `--tags` — comma-separated tag list
### `reply` options
- `topic_id` (positional, **required**) — the topic to reply in
- `-b, --body` **required** (raw markdown/text)
- `-r, --reply-to <post_number>` — reply to a specific **post number** (not id)
### `update` / `delete`
- Both operate on a **post id** (positional, required), not a topic id or
post number. Find the post id via `show` or the raw API (see gotcha below).
- `update -b, --body` **required** (new raw body).
### `search`
- `<query>` (positional, required)
- `-p, --page <n>` — page number (default 1)
### `notifications`
- `-l, --limit <n>` — max results (default 20)
## Key categories
| ID | Slug | Name | Notes |
|----|------|------|-------|
| 4 | `general` | General | Open discussion |
| 23 | `reachableceo` | ReachableCEO | Personal |
| 6 | `chiefoperationsandfinanceofficer` | ChiefOperationsOfficer | COO seat — parent for VP subcategories |
| 3 | `staff` | Staff | May be restricted |
| 74 | `vp-techops` | VP TechOps | ChiefOperationsOfficer (6) — 11 wiki topics migrated from PFVCluster |
| 75 | `vp-compliance` | VP Compliance | ChiefOperationsOfficer (6) — awaiting content |
| 76 | `board` | Board | (top-level, future) |
Run `discourse categories` for the full, current list.
### Key topics (VP TechOps — category 74)
Eleven wiki topics migrated from PFVCluster (Discourse is source of truth):
| Topic | Title | Pattern |
|-------|-------|---------|
| 296 | PFVCluster Project Overview | Pinned wiki |
| 297 | Operations Status | Pinned wiki (updated in place) |
| 298 | Infrastructure Audit Log | Wiki index + dated replies |
| 299 | Network Topology | Wiki |
| 300 | Storage Architecture | Wiki |
| 301 | Data Center Infrastructure | Wiki |
| 302 | Automation and Provisioning | Wiki |
| 303 | Security Architecture and Hardening | Wiki |
| 304 | Proxmox Fleet Reference | Wiki + replies |
| 305 | Kubernetes Platform | Wiki + replies |
| 306 | DNS and DHCP Services | Wiki + replies |
## Gotcha: post id vs post number
Discourse distinguishes a **post number** (1, 2, 3... within a topic — the
first post is always #1) from a **post id** (a globally unique integer).
- `reply -r` takes a **post number**.
- `update` / `delete` take a **post id**.
`show` prints headers like `--- #3 [3] author ...` where the value in brackets
is the post number (here they often coincide for simple topics, but they are
**not** the same thing). To reliably get a post's **id**, either read the raw
JSON via the escape hatch below, or note that `create`/`reply` print the id
of the post they just made (`Posted reply #<num> (id=<id>)`).
## Gotcha: not an admin (user key)
The default API user is trust level 4 (Leader) but **not** an admin/staff
member. With the user-level key:
- **Cannot** create categories, set wiki posts, configure site settings, or
access `/admin/...` endpoints (all 403).
- **Can** create topics, reply, edit own posts, search, list notifications.
For admin operations (category creation, wiki flagging, docs plugin config),
an **admin-scoped API key** is needed. Set it via `DISCOURSE_ADMIN_KEY` in the
credential file and pass it explicitly in raw `requests` calls.
Things that will fail or be restricted without admin:
- Deleting other users' posts.
- Creating topics in staff-only or restricted categories.
- Moving/merging/recategorizing topics.
- Setting the wiki flag on posts.
If an operation 403s, that's expected — surface it to the user rather than
retrying.
## Gotcha: bulk operations & raw API
For anything beyond a single `create`/`reply`/`update`, or to read fields the
CLI doesn't print (e.g. exact post ids, category permissions, user lists),
run Python directly inside the container with `requests`. The CLI uses raw
`requests` against the Discourse REST API (no heavyweight SDK).
Pattern (mount a script and run it in the same image):
```bash
cat > /tmp/script.py <<'PY'
import os, requests
URL = os.environ["DISCOURSE_URL"].rstrip("/")
H = {
"Api-Key": os.environ["DISCOURSE_API_KEY"],
"Api-Username": os.environ["DISCOURSE_API_USERNAME"],
"Accept": "application/json",
}
# Example: get post ids for topic 42
r = requests.get(f"{URL}/t/42.json", headers=H, timeout=30)
r.raise_for_status()
for p in r.json()["post_stream"]["posts"]:
print(p["post_number"], "id=", p["id"], "by", p["username"])
PY
docker run --rm --env-file ~/.creds/discourse.env \
--entrypoint python \
-v /tmp/script.py:/tmp/script.py \
git.knownelement.com/reachableceo/discourse-cli:latest \
/tmp/script.py
```
Useful raw endpoints:
- `GET /t/<topic_id>.json` — full topic incl. `post_stream.posts[]` (each has
`id`, `post_number`, `username`, `cooked`, `raw`).
- `GET /categories.json` — all categories.
- `GET /c/<cat_id>.json` — topics in a category.
- `POST /posts.json` — create topic (`title`+`raw`+`category`) or reply
(`topic_id`+`raw`).
- `PUT /posts/<post_id>.json` — edit (`{"post":{"raw":"..."}}`).
- `DELETE /posts/<post_id>.json` — delete.
- `GET /search.json?q=<query>` — search.
- `GET /notifications.json` — your notifications.
## Patterns
### Pattern: review & respond to a topic
1. `discourse show <topic_id>` — read the topic and all replies.
2. Identify the post you're responding to; note its **post number** (for
`-r`) and the overall context.
3. Draft a reply. `create`/`reply` take **raw markdown** — links, lists,
code fences all work.
4. `discourse reply <topic_id> -b "your markdown" [-r <post_number>]`.
5. Verify with `discourse show <topic_id>`.
### Pattern: before you act
- Always `discourse show <topic_id>` before replying/editing — confirm
the current state so you don't duplicate or contradict prior posts.
- **Never delete a post** unless the user explicitly asks.
- Prefer replying over editing someone else's post (and editing others'
posts will likely 403 anyway as a non-admin).
- When unsure which category to post in, `discourse categories` and pick
the closest match, or ask the user.
### Pattern: posting conventions
- Bodies are **raw markdown** — Discourse renders them. Use fenced code
blocks for commands/output, headings, tables, and bullet lists freely.
- Keep titles concise and descriptive.
- Use tags where the category supports them (`--tags a,b,c`).
### Pattern: wiki topics (living documents)
Wiki topics are the core anti-sprawl primitive. A wiki topic's **first post
is editable by anyone** with permission (not just the original author), and
Discourse preserves the full edit history automatically.
When to use a wiki topic:
- **Living references** — inventories, host lists, network topology, storage
maps. Updated in place as facts change.
- **Collaborative documents** — policies, design docs, specs where multiple
agents/humans contribute. First post = the document; replies = discussion.
- **Status/index pages** — operations status, documentation indexes.
When NOT to use a wiki topic:
- **Discussion threads** — regular topics where each reply is a distinct
voice. Wiki-editing the first post would destroy the conversation.
- **Point-in-time reports** — these go as dated replies inside a wiki
"log" topic (see audit pattern below).
To mark a post as wiki via the API (requires admin/moderator):
```python
requests.put(f"{URL}/posts/{post_id}.json", headers=H, json={"wiki": True})
```
### Pattern: post lifecycle (comment vs replace vs new)
The golden rule: **never create a new topic for an update to existing
knowledge.** One wiki topic per subsystem, updated in place.
| Change type | Action | Why |
|-------------|--------|-----|
| Fact update (new VM, IP changed) | Edit the wiki post in place | Edit history preserves old state |
| New snapshot (audit, drift report) | New reply in the existing log topic | One topic accumulates history |
| Question/discussion about content | Reply to the relevant post | Threaded, doesn't mutate the doc |
| Major restructure | Edit wiki + reply noting why | Edit log = what; reply = why |
### Pattern: audit/snapshot logs
Point-in-time reports (audits, drift reports, status snapshots) do NOT each
get their own topic. Instead, create ONE wiki "log" topic per domain and
append each snapshot as a dated reply:
```
[Wiki post #1] Index table (date | scope | link | superseded-by)
+ latest snapshot summary
[Reply #2] Audit 2026-07-29 — full content
[Reply #3] Audit 2026-07-30 — full content (supersedes #2)
[Reply #4] Audit 2026-08-05 — full content (current)
```
When a new snapshot arrives: add a reply with full content, then edit the
wiki first post to point at the latest reply as "current."
### Pattern: tag taxonomy
Tags are cross-cutting classifiers that prevent category multiplication:
| Tag | Meaning |
|-----|---------|
| `reference` | Living reference doc (inventories, topology, host lists) |
| `runbook` | Operational procedure (deploy, recover, configure) |
| `architecture` | Design doc, system architecture, capacity model |
| `policy` | Naming conventions, security policies, standards |
| `decision` | Resolved decision (ADR, distro choice, analysis outcome) |
| `audit` | Point-in-time snapshot |
| `security` | Security/hardening topic |
### Pattern: category taxonomy
Org-based hierarchy mirroring Known Element's VP seats:
```
ChiefOperationsOfficer (id 6)
├── vp-techops — infrastructure, network, compute, security ops
└── vp-compliance — frameworks, evidence, audit response
Board (future)
```
Security operations (SecOps) lives under vp-techops (security is operational).
Compliance covers frameworks (CMMC/STIG), evidence, and audit response.
## Git workflow
### Atomic commits
Each commit is **one logical change** — one feature, one fix, one doc update.
If you're tempted to write "and also..." in a commit message, that's a sign
to split it into two commits. Stage precisely (`git add <file>`, not
`git add -A`) so unrelated changes don't get bundled.
### Conventional commit messages
Use the [Conventional Commits](https://www.conventionalcommits.org/) format:
```
<type>(<optional scope>): <imperative subject>
<optional body — why, not what>
```
Types used in this repo: `feat`, `fix`, `docs`, `refactor`, `chore`, `style`.
Rules:
- Subject line **under 72 chars**, lowercase, imperative mood.
- No period at end of subject.
- Body wrapped at 72 chars, explains **why** the change exists.
### Commit cadence
- **Commit early and often.** Don't accumulate a pile of unrelated changes.
- Commit **without asking** — if you made a coherent change, commit it.
- Every task (topic review, reply, script addition, doc update) ends with
the relevant files committed.
- Run `git status` before committing to stage only what belongs together.
+16
View File
@@ -0,0 +1,16 @@
FROM python:3.12-slim
ENV PYTHONUNBUFFERED=1 \
PYTHONDONTWRITEBYTECODE=1 \
PIP_NO_CACHE_DIR=1
WORKDIR /app
COPY requirements.txt .
RUN pip install -r requirements.txt
COPY src/discourse_cli.py /usr/local/bin/discourse-cli
RUN chmod +x /usr/local/bin/discourse-cli
ENTRYPOINT ["discourse-cli"]
CMD ["--help"]
+117
View File
@@ -0,0 +1,117 @@
# discourse-cli
A Docker container that lets an AI agent (or a human) read, post, reply,
search, and discuss on a [Discourse](https://www.discourse.org/) forum through
the Discourse REST API. A small Python CLI built on `requests`.
## Requirements
- Docker on the host.
- A Discourse instance with API access enabled
(Admin → API → "All users" or "Single user" API key).
- An API key and the username the key acts as.
## Configuration
Credentials live in the **centralized credential store** at
`~/.creds/discourse.env` (see `tooling-cli/KNELCredsManager`). It holds:
```
DISCOURSE_URL=https://discourse.example.com
DISCOURSE_API_KEY=abc123...
DISCOURSE_API_USERNAME=your-username
```
Use `.env.example` in this directory as a template if you need to create one.
Permissions: directory `700`, the env file `600` (owner read/write only).
## Build
```bash
docker build -t kneldevstack-aimiddleware-discourse-cli:latest .
```
A prebuilt image is also available in the Gitea registry:
```
git.knownelement.com/reachableceo/discourse-cli:latest
```
## Usage
Invoke the container directly with `docker run`. Pass credentials from the
centralized store via `--env-file`:
```bash
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest whoami
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest ls
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest show 42
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest create -c 5 -t "Hello" -b "Body text"
docker run --rm --env-file ~/.creds/discourse.env \
git.knownelement.com/reachableceo/discourse-cli:latest reply 42 -b "Nice point"
```
To use a locally built image instead of the registry image, either set
`DISCOURSE_CLI_IMAGE` or swap the image tag for `kneldevstack-aimiddleware-discourse-cli:latest`.
## Commands
| Command | Description |
| -------------------------------- | ------------------------------------------------------- |
| `whoami` | Show the authenticated user (also a connection test). |
| `categories` | List categories (id, slug, topic count). |
| `cat-info <cat>` | Show details of a category (id or slug). |
| `topics` (`ls`) | List topics (latest, or filtered by category/new/unread).|
| `show <topic_id>` | Show a topic with all its posts. |
| `create` | Create a topic (`--category`, `--title`, `--body`, `--tags`). |
| `reply <topic_id>` | Reply to a topic (`--body`, optional `--reply-to`). |
| `update <post_id>` | Update a post (`--body`). |
| `delete <post_id>` | Delete a post. |
| `search <query>` | Search the forum. |
| `notifications` | List your notifications. |
### `topics` / `ls` filters
- `-c, --category <id|slug>` — only topics in this category
- `-n, --new` — only new (unread tracking) topics
- `-u, --unread` — only unread topics
- `-p, --page <n>` — page number (default 0)
### `create` options
- `-c, --category <id|slug>` — category to post in
- `-t, --title <text>` — topic title (required)
- `-b, --body <text>` — post body, raw text or markdown (required)
- `--tags <a,b,c>` — comma-separated tags
### `reply` options
- `-b, --body <text>` — reply body, raw text or markdown (required)
- `-r, --reply-to <post_number>` — reply to a specific post number (optional)
## Environment variables
| Variable | Required | Description |
| ----------------------- | -------- | ---------------------------------------- |
| `DISCOURSE_URL` | yes | Base URL of the Discourse instance. |
| `DISCOURSE_API_KEY` | yes | API key ("All users" or "Single user"). |
| `DISCOURSE_API_USERNAME`| yes | Username the API key acts as. |
## Validation
```bash
./validate.sh
```
Runs a full live read/write cycle against the configured instance (credential
check, whoami, categories, topics, show, search, create+reply+update+delete
cleanup). Credentials are read from `~/.creds/discourse.env` by default;
override with `DISCOURSE_ENV_FILE`.
+1
View File
@@ -0,0 +1 @@
requests>=2.31.0
+444
View File
@@ -0,0 +1,444 @@
#!/usr/bin/env python3
"""Discourse CLI - a thin wrapper around the Discourse REST API.
Connection details come from the environment:
DISCOURSE_URL base URL of the Discourse instance (e.g. https://discourse.example.com)
DISCOURSE_API_KEY API key ("All users" or "Single user" key from Admin > API)
DISCOURSE_API_USERNAME username the API key acts as (e.g. system, or your account)
Designed to be run inside the discourse-cli Docker container, but works anywhere
these environment variables are set.
"""
import argparse
import json
import os
import sys
import requests
# --------------------------------------------------------------------------- #
# HTTP client
# --------------------------------------------------------------------------- #
class DiscourseAPI:
"""Minimal authenticated Discourse REST client."""
def __init__(self, url, api_key, api_username):
self.base_url = url.rstrip("/")
self.headers = {
"Api-Key": api_key,
"Api-Username": api_username,
"Accept": "application/json",
}
self.timeout = 30
def _request(self, method, path, **kwargs):
url = f"{self.base_url}/{path.lstrip('/')}"
resp = requests.request(
method, url, headers=self.headers, timeout=self.timeout, **kwargs
)
if resp.status_code == 429:
_err(f"rate limited by Discourse (HTTP 429). Retry later.")
if not resp.ok:
detail = ""
try:
detail = resp.json().get("errors", resp.text[:200])
except Exception:
detail = resp.text[:200]
_err(f"Discourse API error {resp.status_code} for {method} {path}: {detail}")
if resp.status_code == 204 or not resp.content:
return {}
return resp.json()
def get(self, path, params=None):
return self._request("GET", path, params=params)
def post(self, path, data=None):
return self._request("POST", path, json=data)
def put(self, path, data=None):
return self._request("PUT", path, json=data)
def delete(self, path, data=None):
return self._request("DELETE", path, json=data)
# --------------------------------------------------------------------------- #
# Helpers
# --------------------------------------------------------------------------- #
def _client():
"""Build and return an authenticated Discourse client, or exit with help."""
url = os.environ.get("DISCOURSE_URL", "").strip()
key = os.environ.get("DISCOURSE_API_KEY", "").strip()
user = os.environ.get("DISCOURSE_API_USERNAME", "").strip()
missing = [
n
for n, v in (
("DISCOURSE_URL", url),
("DISCOURSE_API_KEY", key),
("DISCOURSE_API_USERNAME", user),
)
if not v
]
if missing:
sys.stderr.write(
"ERROR: missing required environment variable(s): "
+ ", ".join(missing)
+ "\n"
)
sys.exit(2)
return DiscourseAPI(url, key, user)
def _err(msg, code=1):
sys.stderr.write(f"ERROR: {msg}\n")
sys.exit(code)
def _kv(label, value):
"""Format a label/value line, omitting falsy values gracefully."""
if value in (None, "", [], {}):
return None
return f"{label:>14}: {value}"
# --------------------------------------------------------------------------- #
# Commands
# --------------------------------------------------------------------------- #
def cmd_whoami(args):
api = _client()
data = api.get("/session/current.json")
u = data.get("current_user", {})
print(f"id: {u.get('id', '?')}")
print(f"username: {u.get('username', '?')}")
print(f"name: {u.get('name', '?')}")
print(f"admin: {u.get('admin', False)}")
print(f"trust: {u.get('trust_level', '?')}")
print(f"discourse: {os.environ['DISCOURSE_URL']}")
return 0
def cmd_categories(args):
api = _client()
data = api.get("/categories.json", params={"include_subcategories": "true"})
cats = data.get("category_list", {}).get("categories", [])
print(f"{'ID':<6} {'SLUG':<36} {'TOPICS':>7} NAME")
print("-" * 90)
count = 0
for c in cats:
topic_count = c.get("topic_count", 0)
print(f"{c.get('id','?'):<6} {str(c.get('slug','')):<36} {topic_count:>6} {c.get('name','')}")
count += 1
for s in c.get("subcategory_list") or []:
sub_count = s.get("topic_count", 0)
slug = " " + str(s.get("slug", ""))
name = " " + s.get("name", "")
print(f"{s.get('id','?'):<6} {slug:<36} {sub_count:>6} {name}")
count += 1
print(f"\n{count} category(ies)")
return 0
def cmd_topics(args):
api = _client()
if args.category:
cat_id = _resolve_category(api, args.category)
data = api.get(f"/c/{cat_id}.json", params={"page": args.page})
topics = data.get("topic_list", {}).get("topics", [])
else:
path = "/latest.json"
if args.unread:
path = "/unread.json"
elif args.new:
path = "/new.json"
data = api.get(path, params={"page": args.page})
topics = data.get("topic_list", {}).get("topics", [])
print(f"{'ID':<10} {'REPLIES':>8} {'VIEWS':>9} TITLE")
print("-" * 90)
for t in topics:
if t.get("pinned"):
continue
replies = t.get("posts_count", 1) - 1
views = t.get("views", 0)
title = t.get("title", "")
print(f"{t.get('id','?'):<10} {replies:>8} {views:>9} {title}")
print(f"\n{len(topics)} topic(s)")
return 0
def cmd_show(args):
api = _client()
data = api.get(f"/t/{args.topic_id}.json")
print(f"#{data.get('id','?')}: {data.get('title','?')}")
print("=" * 90)
print(_kv("category_id", data.get("category_id")))
print(_kv("views", data.get("views")))
print(_kv("like_count", data.get("like_count")))
print(_kv("posts_count", data.get("posts_count")))
print(_kv("created", data.get("created_at")))
print(_kv("last_posted", data.get("last_posted_at")))
print("")
posts = data.get("post_stream", {}).get("posts", [])
for idx, p in enumerate(posts, start=1):
post_num = p.get("post_number", idx)
author = p.get("username", "?")
created = p.get("created_at", "?")
likes = p.get("actions_summary", [])
like_count = 0
for a in likes:
if a.get("id") == 2:
like_count = a.get("count", 0)
print(f"\n--- #{post_num} [{post_num}] {author} ({created}) likes={like_count} ---")
cooked = p.get("cooked", "")
text = _strip_html(cooked)
if text.strip():
print(text.strip())
else:
print("(no content)")
return 0
def cmd_create(args):
api = _client()
if not args.title:
_err("--title is required to create a topic")
if not args.body:
_err("--body is required to create a topic")
cat_id = _resolve_category(api, args.category) if args.category else None
payload = {
"title": args.title,
"raw": args.body,
}
if cat_id is not None:
payload["category"] = cat_id
if args.tags:
payload["tags"] = [t.strip() for t in args.tags.split(",") if t.strip()]
data = api.post("/posts.json", data=payload)
topic_id = data.get("topic_id", "?")
post_id = data.get("id", "?")
print(f"Created topic #{topic_id} (post #{post_id}): {args.title}")
if topic_id != "?":
print(f"URL: {os.environ['DISCOURSE_URL']}/t/{topic_id}")
return 0
def cmd_reply(args):
api = _client()
if not args.body:
_err("--body is required to reply")
payload = {"topic_id": args.topic_id, "raw": args.body}
if args.reply_to:
payload["reply_to_post_number"] = args.reply_to
data = api.post("/posts.json", data=payload)
post_id = data.get("id", "?")
post_num = data.get("post_number", "?")
print(f"Posted reply #{post_num} (id={post_id}) in topic #{args.topic_id}")
return 0
def cmd_update(args):
api = _client()
if not args.body:
_err("--body is required to update a post")
data = api.put(f"/posts/{args.post_id}.json", data={"post": {"raw": args.body}})
post_num = data.get("post", {}).get("post_number", "?")
print(f"Updated post #{post_num} (id={args.post_id})")
return 0
def cmd_delete(args):
api = _client()
api.delete(f"/posts/{args.post_id}.json")
print(f"Deleted post #{args.post_id}")
return 0
def cmd_search(args):
api = _client()
params = {"q": args.query, "page": args.page}
data = api.get("/search.json", params=params)
topics = data.get("topics", [])
posts = data.get("posts", [])
topic_map = {t.get("id"): t for t in topics}
print(f"{'TOPIC':<10} {'POST':<8} BLURB")
print("-" * 90)
for p in posts:
tid = p.get("topic_id", "?")
t = topic_map.get(tid, {})
title = t.get("title", "")
blurb = p.get("blurb", "")
pid = p.get("id", "?")
print(f"{tid:<10} {pid:<8} {title}")
if blurb:
print(f"{'':<20}{blurb}")
print(f"\n{len(posts)} result(s)")
return 0
def cmd_notifications(args):
api = _client()
data = api.get("/notifications.json")
notifs = data.get("notifications", [])
if not notifs:
print("(no notifications)")
return 0
print(f"{'ID':<10} {'READ':>5} TYPE SUBJECT")
print("-" * 90)
for n in notifs[: args.limit]:
nid = n.get("id", "?")
read = "yes" if n.get("read") else "no"
ntype = n.get("notification_type", "?")
subject = n.get("slug") or n.get("data", {}).get("topic_title", "")
print(f"{nid:<10} {read:>5} {str(ntype):<20} {subject}")
print(f"\n{len(notifs)} notification(s), showing {min(len(notifs), args.limit)}")
return 0
def cmd_categories_info(args):
api = _client()
cat_id = _resolve_category(api, args.category)
data = api.get(f"/c/{cat_id}/show.json")
c = data.get("category", {})
print(f"#{c.get('id','?')}: {c.get('name','?')}")
print("=" * 60)
print(_kv("slug", c.get("slug")))
print(_kv("color", c.get("color")))
print(_kv("topic_count", c.get("topic_count")))
print(_kv("post_count", c.get("post_count")))
print(_kv("description", _strip_html(c.get("description", "") or "")[:200]))
return 0
# --------------------------------------------------------------------------- #
# Resolution helpers
# --------------------------------------------------------------------------- #
def _resolve_category(api, ref):
"""Resolve a category id, slug, or name to a numeric id."""
if str(ref).isdigit():
return int(ref)
data = api.get("/categories.json", params={"include_subcategories": "true"})
cats = data.get("category_list", {}).get("categories", [])
needle = str(ref).strip().lower()
for c in cats:
if str(c.get("slug", "")).lower() == needle:
return c["id"]
if str(c.get("name", "")).lower() == needle:
return c["id"]
for s in c.get("subcategory_list") or []:
if str(s.get("slug", "")).lower() == needle:
return s["id"]
if str(s.get("name", "")).lower() == needle:
return s["id"]
_err(f"unknown category '{ref}'. Run 'categories' to list valid slugs/names.")
def _strip_html(cooked):
"""Very small HTML-to-text for display of Discourse 'cooked' post bodies."""
import html
import re
if not cooked:
return ""
# Preserve block-level breaks
text = re.sub(r"(?i)</(p|div|li|h[1-6]|tr|blockquote)>", "\n", cooked)
text = re.sub(r"(?i)<br\s*/?>", "\n", text)
text = re.sub(r"(?i)<li[^>]*>", " - ", text)
# Code blocks
text = re.sub(r"(?i)<code[^>]*>", "`", text)
text = re.sub(r"(?i)</code>", "`", text)
# Strip all remaining tags
text = re.sub(r"<[^>]+>", "", text)
text = html.unescape(text)
# Collapse excessive blank lines
text = re.sub(r"\n{3,}", "\n\n", text)
return text
# --------------------------------------------------------------------------- #
# Argument parsing
# --------------------------------------------------------------------------- #
def build_parser():
p = argparse.ArgumentParser(
prog="discourse-cli",
description="Read, post, and discuss on a Discourse forum via the REST API.",
)
sub = p.add_subparsers(dest="command", required=True)
sub.add_parser(
"whoami", help="show the authenticated user (connection test)"
).set_defaults(func=cmd_whoami)
sp = sub.add_parser("categories", help="list categories")
sp.set_defaults(func=cmd_categories)
sp = sub.add_parser("cat-info", help="show details of a category")
sp.add_argument("category", metavar="ID_OR_SLUG")
sp.set_defaults(func=cmd_categories_info)
sp = sub.add_parser("topics", aliases=["ls"], help="list topics (latest or in a category)")
sp.add_argument("-c", "--category", metavar="ID_OR_SLUG", help="filter by category")
sp.add_argument("-n", "--new", action="store_true", help="only new topics")
sp.add_argument("-u", "--unread", action="store_true", help="only unread topics")
sp.add_argument("-p", "--page", type=int, default=0, help="page number (default 0)")
sp.set_defaults(func=cmd_topics)
sp = sub.add_parser("show", help="show a topic with its posts")
sp.add_argument("topic_id", type=int)
sp.set_defaults(func=cmd_show)
sp = sub.add_parser("create", help="create a new topic")
sp.add_argument("-c", "--category", metavar="ID_OR_SLUG")
sp.add_argument("-t", "--title", required=True, help="topic title")
sp.add_argument("-b", "--body", required=True, help="post body (raw text/markdown)")
sp.add_argument("--tags", help="comma-separated tags")
sp.set_defaults(func=cmd_create)
sp = sub.add_parser("reply", help="reply to a topic")
sp.add_argument("topic_id", type=int)
sp.add_argument("-b", "--body", required=True, help="reply body (raw text/markdown)")
sp.add_argument(
"-r", "--reply-to", type=int, metavar="POST_NUMBER", help="reply to a specific post number"
)
sp.set_defaults(func=cmd_reply)
sp = sub.add_parser("update", help="update an existing post")
sp.add_argument("post_id", type=int)
sp.add_argument("-b", "--body", required=True, help="new post body (raw text/markdown)")
sp.set_defaults(func=cmd_update)
sp = sub.add_parser("delete", help="delete a post")
sp.add_argument("post_id", type=int)
sp.set_defaults(func=cmd_delete)
sp = sub.add_parser("search", help="search the forum")
sp.add_argument("query", help="search query")
sp.add_argument("-p", "--page", type=int, default=1, help="page number (default 1)")
sp.set_defaults(func=cmd_search)
sp = sub.add_parser("notifications", help="list your notifications")
sp.add_argument("-l", "--limit", type=int, default=20, help="max results (default 20)")
sp.set_defaults(func=cmd_notifications)
return p
def main(argv=None):
parser = build_parser()
args = parser.parse_args(argv)
try:
return args.func(args)
except requests.exceptions.ConnectionError:
_err(f"could not connect to Discourse at {os.environ.get('DISCOURSE_URL','?')}", 2)
except requests.exceptions.Timeout:
_err("request timed out talking to Discourse", 2)
except SystemExit:
raise
except Exception as e: # noqa: BLE001 - top-level safety net for the CLI
_err(f"{type(e).__name__}: {e}")
if __name__ == "__main__":
sys.exit(main())
+210
View File
@@ -0,0 +1,210 @@
#!/bin/sh
# discourse-cli validation script.
# Runs everything inside containers so no host tooling (curl, etc.) is needed.
#
# Usage: ./validate.sh
#
# Exit codes: 0 = all checks passed, 1 = one or more checks failed
set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
cd "$SCRIPT_DIR"
# Credentials live in the centralized ~/.creds/ store (see KNELCredsManager).
# Override with DISCOURSE_ENV_FILE if you need a different location.
ENV_FILE="${DISCOURSE_ENV_FILE:-$HOME/.creds/discourse.env}"
# Ensure env file exists
if [ ! -f "$ENV_FILE" ]; then
echo "FAIL: env file not found: $ENV_FILE"
echo " Set DISCOURSE_ENV_FILE or create it in ~/.creds/discourse.env"
exit 1
fi
IMAGE="${DISCOURSE_CLI_IMAGE:-git.knownelement.com/reachableceo/discourse-cli:latest}"
PASS=0
FAIL=0
TOTAL=0
ok() { TOTAL=$((TOTAL+1)); echo "PASS: $1"; PASS=$((PASS+1)); }
fail() { TOTAL=$((TOTAL+1)); echo "FAIL: $1"; FAIL=$((FAIL+1)); }
echo "============================================"
echo " discourse-cli validation"
echo " image: ${IMAGE}"
echo "============================================"
echo ""
# ---------------------------------------------------------------- #
# 0. Credential sanity (raw HTTP via containerized curl)
# ---------------------------------------------------------------- #
echo "--- [0] Credential check (raw HTTP via containerized curl) ---"
# Load env vars from the creds file for the raw test
DISCOURSE_URL=$(grep -E '^DISCOURSE_URL=' "$ENV_FILE" | cut -d= -f2-)
DISCOURSE_API_KEY=$(grep -E '^DISCOURSE_API_KEY=' "$ENV_FILE" | cut -d= -f2-)
DISCOURSE_API_USERNAME=$(grep -E '^DISCOURSE_API_USERNAME=' "$ENV_FILE" | cut -d= -f2-)
HTTP_CODE=$(docker run --rm curlimages/curl:latest \
-s -o /dev/null -w '%{http_code}' \
-H "Api-Key: ${DISCOURSE_API_KEY}" \
-H "Api-Username: ${DISCOURSE_API_USERNAME}" \
-H "Accept: application/json" \
"${DISCOURSE_URL}/session/current.json" 2>/dev/null || echo "000")
if [ "$HTTP_CODE" = "200" ]; then
ok "raw HTTP credentials accepted (HTTP ${HTTP_CODE})"
else
fail "raw HTTP credentials rejected (HTTP ${HTTP_CODE})"
echo " -> Discourse at ${DISCOURSE_URL} returned ${HTTP_CODE} for /session/current.json"
echo " -> Check: API key active? IP allowlist? Username matches key scope?"
echo ""
echo " Stopping: cannot validate CLI commands without valid credentials."
echo ""
echo "============================================"
echo " RESULTS: ${PASS}/${TOTAL} passed, ${FAIL} failed"
echo "============================================"
exit 1
fi
echo ""
# ---------------------------------------------------------------- #
# 1. whoami (live connection test)
# ---------------------------------------------------------------- #
echo "--- [1] whoami ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" whoami 2>&1) || true
if echo "$OUT" | grep -qE '^username:'; then
ok "whoami returned user info"
echo " $OUT" | head -5
else
fail "whoami did not return expected output"
echo " $OUT" | head -5
fi
echo ""
# ---------------------------------------------------------------- #
# 2. categories (live read)
# ---------------------------------------------------------------- #
echo "--- [2] categories ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" categories 2>&1) || true
if echo "$OUT" | grep -qE '^[0-9]+ category'; then
ok "categories returned data"
echo " $OUT" | head -6
# Grab first category id for later tests
CAT_ID=$(echo "$OUT" | grep -E '^[0-9]' | head -1 | awk '{print $1}')
else
fail "categories did not return expected output"
echo " $OUT" | head -6
CAT_ID=""
fi
echo ""
# ---------------------------------------------------------------- #
# 3. topics / ls (live read)
# ---------------------------------------------------------------- #
echo "--- [3] topics (latest) ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" ls 2>&1) || true
if echo "$OUT" | grep -qE 'topic\(s\)'; then
ok "topics list returned data"
echo " $OUT" | head -6
TOPIC_ID=$(echo "$OUT" | grep -E '^[0-9]' | head -1 | awk '{print $1}')
else
fail "topics list did not return expected output"
echo " $OUT" | head -6
TOPIC_ID=""
fi
echo ""
# ---------------------------------------------------------------- #
# 4. show topic (live read)
# ---------------------------------------------------------------- #
if [ -n "${TOPIC_ID:-}" ]; then
echo "--- [4] show topic ${TOPIC_ID} ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" show "${TOPIC_ID}" 2>&1) || true
if echo "$OUT" | grep -qE '^#'; then
ok "show topic ${TOPIC_ID} returned content"
echo " $OUT" | head -8
else
fail "show topic ${TOPIC_ID} did not return expected output"
echo " $OUT" | head -8
fi
else
echo "--- [4] show topic (SKIPPED - no topic id found in step 3) ---"
fi
echo ""
# ---------------------------------------------------------------- #
# 5. search (live read)
# ---------------------------------------------------------------- #
echo "--- [5] search ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" search "test" 2>&1) || true
if echo "$OUT" | grep -qE 'result\(s\)'; then
ok "search returned results"
echo " $OUT" | head -5
else
fail "search did not return expected output"
echo " $OUT" | head -5
fi
echo ""
# ---------------------------------------------------------------- #
# 6. create + reply + update + delete (live write cycle)
# ---------------------------------------------------------------- #
echo "--- [6] create + reply + update + delete (write cycle) ---"
CREATE_OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" create \
${CAT_ID:+-c "$CAT_ID"} \
-t "discourse-cli validation test $(date +%s)" \
-b "This is an automated validation test post. It will be cleaned up shortly after creation." \
2>&1) || true
NEW_TOPIC_ID=$(echo "$CREATE_OUT" | grep -oE 'topic #[0-9]+' | grep -oE '[0-9]+')
if [ -n "${NEW_TOPIC_ID:-}" ]; then
ok "create topic succeeded (topic #${NEW_TOPIC_ID})"
echo " $CREATE_OUT"
# reply
REPLY_OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" reply "${NEW_TOPIC_ID}" \
-b "Validation reply test." 2>&1) || true
POST_ID=$(echo "$REPLY_OUT" | grep -oE 'id=[0-9]+' | grep -oE '[0-9]+')
if echo "$REPLY_OUT" | grep -qE 'Posted reply'; then
ok "reply succeeded"
echo " $REPLY_OUT"
else
fail "reply failed"
echo " $REPLY_OUT"
fi
# update
if [ -n "${POST_ID:-}" ]; then
UPD_OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" update "${POST_ID}" \
-b "Updated validation reply test." 2>&1) || true
if echo "$UPD_OUT" | grep -qE 'Updated post'; then
ok "update post ${POST_ID} succeeded"
else
fail "update post ${POST_ID} failed"
echo " $UPD_OUT"
fi
# delete
DEL_OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" delete "${POST_ID}" 2>&1) || true
if echo "$DEL_OUT" | grep -qE 'Deleted post'; then
ok "delete post ${POST_ID} succeeded"
else
fail "delete post ${POST_ID} failed"
echo " $DEL_OUT"
fi
fi
else
fail "create topic failed"
echo " $CREATE_OUT"
fi
echo ""
# ---------------------------------------------------------------- #
# Results
# ---------------------------------------------------------------- #
echo "============================================"
echo " RESULTS: ${PASS}/${TOTAL} passed, ${FAIL} failed"
echo "============================================"
[ "$FAIL" -eq 0 ]
+5
View File
@@ -0,0 +1,5 @@
# Technitium DNS connection details
# Copy this file to .env and fill in real values. The .env file is gitignored.
TECHNITIUM_URL=http://pfv-netinfra-01:5380
TECHNITIUM_DNS_TOKEN=your-dns-api-token-here
TECHNITIUM_DNS_ZONE=knel.net
+8
View File
@@ -0,0 +1,8 @@
# Secrets - never commit
.env
# OS / editor cruft
.DS_Store
*.swp
*.swo
*~
+97
View File
@@ -0,0 +1,97 @@
# AGENTS.md — DNS CLI tooling
This directory holds the `dns-cli` Docker container — a bash CLI over the
Technitium DNS Server REST API. Every Crush session that works on or with
this tool should read this file first.
## Connection & invocation
Invoke the real container with `docker run`. Credentials come from the
centralized credential store (see `tooling-cli/KNELCredsManager`):
```bash
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest <command>
```
There is intentionally **no `bin/` wrapper** — invoke the real container, as
the project-wide rules require. For readability in an interactive session you
may define a throwaway shell alias (do not commit one):
```bash
alias dns='docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest'
```
The examples below write `dns <cmd>` for brevity; expand it to the full
`docker run` line (or set the alias) before running.
| Item | Value |
|------|-------|
| Instance | `http://pfv-netinfra-01.knel.net:5380` |
| Default zone | `knel.net` |
| Credentials | `~/.creds/technitium.env` (`TECHNITIUM_URL`, `TECHNITIUM_DNS_TOKEN`, `TECHNITIUM_DNS_ZONE`) |
| CLI image | `git.knownelement.com/reachableceo/dns-cli:latest` |
| Source | `src/dns_cli.sh` in this directory |
## Quick start
```bash
# List all zones:
dns zones
# List records in the default zone:
dns list
# Search for a record:
dns search ns1
# Resolve a name:
dns get pfv-netinfra-01
```
## Command reference
| Command | Description |
|---------|-------------|
| `zones` | List all zones. |
| `list [zone]` | List records in a zone (default: `TECHNITIUM_DNS_ZONE`). |
| `get <name>` | Resolve a record via `dig` (A/AAAA). |
| `add <name> <ip> [ttl]` | Add an A record with auto-PTR (default TTL 3600). |
| `delete <name> <ip>` | Delete an A record. |
| `flush` | Flush the DNS cache. |
| `search <pattern>` | Search records by name pattern (case-insensitive). |
## Patterns
### Pattern: add a new DNS record
1. `dns search <hostname>` — check if the name already exists.
2. `dns add <hostname> <ip>` — creates the A record with auto-PTR.
3. `dns get <hostname>` — verify resolution.
4. `dns flush` — flush cache so the new record is immediately resolvable.
### Pattern: before you act
- Always `dns search <pattern>` or `dns list` before adding/deleting —
confirm current state so you don't create duplicates or delete the wrong
record.
- **Never delete a record** unless the user explicitly asks.
- PTR records are auto-created on `add` and auto-deleted on `delete`.
## Git workflow
### Atomic commits
Each commit is **one logical change**. Stage precisely (`git add <file>`,
not `git add -A`).
### Conventional commit messages
```
<type>(<optional scope>): <imperative subject>
<optional body — why, not what>
```
Types: `feat`, `fix`, `docs`, `refactor`, `chore`, `style`.
- Subject line under 72 chars, lowercase, imperative mood.
- No period at end of subject.
- Body wrapped at 72 chars, explains **why** the change exists.
+9
View File
@@ -0,0 +1,9 @@
FROM alpine:3.20
RUN apk add --no-cache bash curl python3 bind-tools
COPY src/dns_cli.sh /usr/local/bin/dns-cli
RUN chmod +x /usr/local/bin/dns-cli
ENTRYPOINT ["dns-cli"]
CMD ["--help"]
+82
View File
@@ -0,0 +1,82 @@
# dns-cli
A Docker container that lets an AI agent (or a human) manage DNS records on a
[Technitium DNS Server](https://technitium.com/dns/) instance through its REST
API. A small bash CLI built on `curl` + `python3` + `dig`.
## Requirements
- Docker on the host.
- A Technitium DNS Server instance with the REST API enabled.
- A DNS API token (Technitium → Settings → API Token).
## Configuration
Credentials live in the **centralized credential store** at
`~/.creds/technitium.env` (see `tooling-cli/KNELCredsManager`). It holds:
```
TECHNITIUM_URL=http://pfv-netinfra-01:5380
TECHNITIUM_DNS_TOKEN=your-api-token
TECHNITIUM_DNS_ZONE=knel.net
```
Use `.env.example` in this directory as a template if you need to create one.
Permissions: directory `700`, the env file `600` (owner read/write only).
## Build
```bash
docker build -t git.knownelement.com/reachableceo/dns-cli:latest .
```
The image is also available in the Gitea registry:
```
git.knownelement.com/reachableceo/dns-cli:latest
```
## Usage
Invoke the container directly with `docker run`. Pass credentials from the
centralized store via `--env-file`:
```bash
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest zones
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest list
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest search ns1
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest add myhost 192.168.1.50
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest delete myhost 192.168.1.50
docker run --rm --env-file ~/.creds/technitium.env \
git.knownelement.com/reachableceo/dns-cli:latest flush
```
## Commands
| Command | Description |
|---------|-------------|
| `zones` | List all zones. |
| `list [zone]` | List records in a zone (default: `TECHNITIUM_DNS_ZONE`). |
| `get <name>` | Resolve a record via `dig` (A/AAAA). |
| `add <name> <ip> [ttl]` | Add an A record with auto-PTR (default TTL 3600). |
| `delete <name> <ip>` | Delete an A record. |
| `flush` | Flush the DNS cache. |
| `search <pattern>` | Search records by name pattern (case-insensitive). |
## Environment variables
| Variable | Required | Description |
|----------|----------|-------------|
| `TECHNITIUM_URL` | yes | Base URL of the Technitium instance. |
| `TECHNITIUM_DNS_TOKEN` | yes | API token for the DNS REST API. |
| `TECHNITIUM_DNS_ZONE` | yes | Default zone for add/delete/search operations. |
+125
View File
@@ -0,0 +1,125 @@
#!/usr/bin/env bash
# dns-cli - CLI wrapper for Technitium DNS Server API
#
# Usage:
# dns-cli zones List all zones
# dns-cli list [zone] List records in a zone (default: $TECHNITIUM_DNS_ZONE)
# dns-cli get <name> Resolve a record (A/AAAA)
# dns-cli add <name> <ip> [ttl] Add A record (+ auto PTR)
# dns-cli delete <name> <ip> Delete A record
# dns-cli flush Flush DNS cache
# dns-cli search <pattern> Search records by name pattern
#
# Env (provided via --env-file ~/.creds/technitium.env):
# TECHNITIUM_URL Technitium base URL (required)
# TECHNITIUM_DNS_TOKEN API token (required)
# TECHNITIUM_DNS_ZONE Default zone (required)
set -euo pipefail
URL="${TECHNITIUM_URL:?TECHNITIUM_URL is required}"
TOKEN="${TECHNITIUM_DNS_TOKEN:?TECHNITIUM_DNS_TOKEN is required}"
ZONE="${TECHNITIUM_DNS_ZONE:?TECHNITIUM_DNS_ZONE is required}"
# API helper — adds Bearer auth, returns JSON
api() {
local method="$1" endpoint="$2"; shift 2
local args=()
for kv in "$@"; do
args+=(-d "$kv")
done
curl -s -H "Authorization: Bearer $TOKEN" -X "$method" "${args[@]}" \
"$URL/api/$endpoint" 2>/dev/null
}
# Subcommands
cmd_zones() {
api GET "zones/list" | python3 -c "
import json,sys
d=json.load(sys.stdin)
for z in d.get('response',{}).get('zones',[]):
print(z['name'])
"
}
cmd_list() {
local zone="${1:-$ZONE}"
api GET "zones/records/get?domain=$zone&zone=$zone&listZone=true" | python3 -c "
import json,sys
d=json.load(sys.stdin)
recs = d.get('response',{}).get('records',[])
for r in sorted(recs, key=lambda x: (x['name'], x['type'])):
rd = r.get('rData',{})
ip = rd.get('ipAddress','')
val = ip or rd.get('nameServer','') or rd.get('primaryNameServer','') or str(rd)
print(f\"{r['name']:50s} {r['type']:6s} {r.get('ttl',''):6} {val}\")
"
}
cmd_get() {
local name="$1"
dig +short "$name" 2>/dev/null || true
}
cmd_add() {
local name="$1" ip="$2" ttl="${3:-3600}"
local fqdn="$name"
[[ "$fqdn" != *.* ]] && fqdn="$name.$ZONE"
api POST "zones/records/add" \
"zone=$ZONE" "domain=$fqdn" "type=A" "ttl=$ttl" "ipAddress=$ip" "ptr=true" "overwrite=true" | python3 -c "
import json,sys
d=json.load(sys.stdin)
print(d.get('status','?'), d.get('errorMessage',''))
"
}
cmd_delete() {
local name="$1" ip="$2"
local fqdn="$name"
[[ "$fqdn" != *.* ]] && fqdn="$name.$ZONE"
api POST "zones/records/delete" \
"zone=$ZONE" "domain=$fqdn" "type=A" "ipAddress=$ip" | python3 -c "
import json,sys
d=json.load(sys.stdin)
print(d.get('status','?'), d.get('errorMessage',''))
"
}
cmd_flush() {
api POST "cache/flush" | python3 -c "
import json,sys
d=json.load(sys.stdin)
print(d.get('status','?'))
"
}
cmd_search() {
local pattern="$1"
api GET "zones/records/get?domain=$ZONE&zone=$ZONE&listZone=true" | python3 -c "
import json,sys
d=json.load(sys.stdin)
recs = d.get('response',{}).get('records',[])
for r in sorted(recs, key=lambda x: x['name']):
if '$pattern' in r['name'].lower():
rd = r.get('rData',{})
ip = rd.get('ipAddress','')
val = ip or rd.get('nameServer','') or str(rd)
print(f\"{r['name']:50s} {r['type']:6s} {val}\")
"
}
# Main
case "${1:-}" in
zones) cmd_zones ;;
list) shift; cmd_list "${1:-}" ;;
get) shift; cmd_get "$1" ;;
add) shift; cmd_add "$@" ;;
delete) shift; cmd_delete "$@" ;;
flush) cmd_flush ;;
search) shift; cmd_search "$1" ;;
""|-h|--help|help)
sed -n '2,18p' "$0" >&2
exit 0
;;
*) echo "Unknown command: $1" >&2; exit 1 ;;
esac
+81
View File
@@ -0,0 +1,81 @@
#!/bin/sh
# dns-cli validation script.
# Runs a live read-only cycle against the configured Technitium instance.
#
# Usage: ./validate.sh
# Exit codes: 0 = all checks passed, 1 = one or more checks failed
set -eu
SCRIPT_DIR="$(cd "$(dirname "$0")" && pwd)"
cd "$SCRIPT_DIR"
ENV_FILE="${TECHNITIUM_ENV_FILE:-$HOME/.creds/technitium.env}"
if [ ! -f "$ENV_FILE" ]; then
echo "FAIL: env file not found: $ENV_FILE"
echo " Set TECHNITIUM_ENV_FILE or create it in ~/.creds/technitium.env"
exit 1
fi
IMAGE="${DNS_CLI_IMAGE:-git.knownelement.com/reachableceo/dns-cli:latest}"
PASS=0
FAIL=0
TOTAL=0
ok() { TOTAL=$((TOTAL+1)); echo "PASS: $1"; PASS=$((PASS+1)); }
fail() { TOTAL=$((TOTAL+1)); echo "FAIL: $1"; FAIL=$((FAIL+1)); }
echo "============================================"
echo " dns-cli validation"
echo " image: ${IMAGE}"
echo "============================================"
echo ""
# ---------------------------------------------------------------- #
# 1. zones (live read)
# ---------------------------------------------------------------- #
echo "--- [1] zones ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" zones 2>&1) || true
if echo "$OUT" | grep -qE '\.'; then
ok "zones returned data"
echo "$OUT" | head -5
else
fail "zones did not return expected output"
echo " $OUT" | head -5
fi
echo ""
# ---------------------------------------------------------------- #
# 2. list (live read)
# ---------------------------------------------------------------- #
echo "--- [2] list ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" list 2>&1) || true
if echo "$OUT" | grep -qE 'A\s'; then
ok "list returned records"
echo "$OUT" | head -5
else
fail "list did not return expected output"
echo " $OUT" | head -5
fi
echo ""
# ---------------------------------------------------------------- #
# 3. search (live read)
# ---------------------------------------------------------------- #
echo "--- [3] search ---"
OUT=$(docker run --rm --env-file "$ENV_FILE" "${IMAGE}" search "ns" 2>&1) || true
if [ -n "$OUT" ]; then
ok "search returned results"
echo "$OUT" | head -3
else
fail "search returned no results"
fi
echo ""
# ---------------------------------------------------------------- #
# Results
# ---------------------------------------------------------------- #
echo "============================================"
echo " RESULTS: ${PASS}/${TOTAL} passed, ${FAIL} failed"
echo "============================================"
[ "$FAIL" -eq 0 ]
+8
View File
@@ -0,0 +1,8 @@
# Secrets - never commit
.env
# OS / editor cruft
.DS_Store
*.swp
*.swo
*~
+240
View File
@@ -0,0 +1,240 @@
# AGENTS.md — Redmine CLI tooling
This directory holds the `redmine-cli` Docker container — a thin Python CLI
over `python-redmine` for Redmine issue tracking. Every Crush session that
works on or with this tool should read this file first: how to invoke it,
what commands exist, what gotchas to avoid, and the patterns to follow for
common ticket operations.
## Connection & invocation
Invoke the real container with `docker run`. Credentials come from the
centralized credential store (see `tooling-cli/KNELCredsManager`):
```bash
docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest <command>
```
There is intentionally **no `bin/` wrapper** — invoke the real container, as
the project-wide rules require. For readability in an interactive session you
may define a throwaway shell alias (do not commit one):
```bash
alias redmine='docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest'
```
The examples below write `redmine <cmd>` for brevity; expand it to the full
`docker run` line (or set the alias) before running.
| Item | Value |
|------|-------|
| Instance | `https://projects.knownelement.com` |
| API user | Charles N (id 5, login `reachableceo`) |
| Credentials | `~/.creds/redmine.env` (`REDMINE_URL`, `REDMINE_API_KEY`) |
| CLI image | `git.knownelement.com/reachableceo/redmine-cli:latest` |
| Source | `src/redmine_cli.py` in this directory |
To use a locally built image instead of the registry image, either set
`REDMINE_CLI_IMAGE` or swap the tag for `kneldevstack-aimiddleware-redmine-cli:latest`.
## Quick start
```bash
# Connection sanity check (run this first in any session):
redmine whoami
# Look at a ticket:
redmine show 314
# Your queue:
redmine list --assigned-to-me
```
## Command reference
| Command | Description |
|---------|-------------|
| `whoami` | Authenticated user + connection test. |
| `projects` | List projects (id, identifier, name). |
| `statuses` | List issue statuses + which are "closed". |
| `list` (`ls`) | List issues. Filters below. |
| `show <id>` | Full issue detail incl. note history. |
| `create` | Create an issue. |
| `update <id>` | Update an issue (status, notes, assignee, done, ...). |
| `close <id>` | Close (first closed status, done ratio 100%). |
### `list` filters
- `-m, --assigned-to-me` — only issues assigned to you
- `-a, --assigned-to <id>` — filter by assignee user id
- `-p, --project <id|slug>` — filter by project
- `-s, --status <name|id|open|closed>` — filter by status
- `-l, --limit <n>` — max results (default 50)
- `--sort <spec>` — Redmine sort spec (default `priority:desc,updated_on:desc`)
### `create` options
- `-p, --project <id|slug>` **required**
- `-s, --subject` **required**
- `-d, --description`
- `-a, --assigned-to <id>`
- `-t, --tracker <id>`
- `--priority <id>`
- `--status <name|id>`
### `update` options
- `-s, --status <name|id>`
- `-n, --notes <text>`
- `-a, --assigned-to <id>`
- `--done-ratio <0-100>`
- `--subject`
- `--priority <id>`
## Key project / tracker / status IDs
### Most-used projects
| ID | Identifier | Name |
|----|------------|------|
| 55 | `technicaloperations` | Known Element Enterprises - Technology & Facility Services |
| 62 | `business-operations` | Known Element Enterprises - Business Services |
| 77 | `tsys-group` | TSYS Group |
Run `redmine projects` for the full list.
### Statuses
| ID | Name | Closed? |
|----|------|---------|
| 1 | New | no |
| 2 | In Progress | no |
| 3 | Resolved | yes |
| 4 | Feedback | no |
| 5 | Closed | yes |
| 6 | Rejected | no |
### Trackers
Tracker 3 = **Support** (the most common one in project 55). Inspect a
parent issue to inherit its exact tracker.
## Gotcha: creating subtasks
`redmine create` has **no `--parent` flag**. To create a subtask of an
existing issue you must use `python-redmine` directly inside the container
(via `--entrypoint python`).
Pattern (mount a script and run it in the same image):
```bash
cat > /tmp/script.py <<'PY'
import os
from redminelib import Redmine
rm = Redmine(os.environ["REDMINE_URL"].rstrip("/"), key=os.environ["REDMINE_API_KEY"])
issue = rm.issue.create(
project_id=55, # inherit from parent
tracker_id=3, # inherit from parent
priority_id=2, # inherit from parent
status_id=4, # Feedback = freshly created, awaiting work
assigned_to_id=5, # inherit from parent
parent_issue_id=314, # THE KEY FIELD
subject="Your subject",
description="Your description",
)
print(f"Created #{issue.id}")
PY
docker run --rm --env-file ~/.creds/redmine.env \
--entrypoint python \
-v /tmp/script.py:/tmp/script.py \
git.knownelement.com/reachableceo/redmine-cli:latest \
/tmp/script.py
```
**Before creating subtasks**, inspect the parent to inherit its attributes:
```bash
docker run --rm --env-file ~/.creds/redmine.env \
--entrypoint python \
git.knownelement.com/reachableceo/redmine-cli:latest -c "
import os
from redminelib import Redmine
rm = Redmine(os.environ['REDMINE_URL'].rstrip('/'), key=os.environ['REDMINE_API_KEY'])
i = rm.issue.get(PARENT_ID)
print('project_id:', i.project.id)
print('tracker_id:', i.tracker.id, i.tracker.name)
print('priority_id:', i.priority.id, i.priority.name)
print('assigned_to_id:', getattr(i.assigned_to,'id',None))
print('status_id:', i.status.id, i.status.name)
"
```
## Patterns
### Pattern: review & split a ticket
1. `redmine show <id>` — read the full ticket (description + history).
2. Identify the **distinct bodies of work**. Each should be independently
trackable and assignable.
3. Map out **dependencies** between the pieces (what blocks what).
4. Propose a split to the user as a table: proposed subject, scope, and
dependencies. Wait for approval before creating anything.
5. On approval: create subtasks (see "Gotcha: creating subtasks" above),
inheriting project/tracker/priority/assignee from the parent.
6. Update the parent's **description** to an index table of children +
recommended execution order. Add a note explaining the split.
7. Verify with `redmine show <parent_id>`.
### Pattern: ticket conventions
- **Subject prefix:** OAM-pool tickets use `OAM: <topic>`. Match the parent's
naming convention if one exists.
- **Description header:** always start with `Parent/umbrella: #<id> (<subject>).`
- **Scope section:** bullet list of concrete deliverables. Use `- [ ]` for
checklist items within a subtask.
- **Dependencies section:** list what the ticket depends on and what it blocks,
referencing ticket numbers once they exist.
- **Newly created children** go to status **Feedback (4)** so they're visible
but not yet "in progress".
- **Tables** in descriptions render in Redmine's Markdown pipeline (`| a | b |`).
### Pattern: bulk operations
For anything beyond a single `create`/`update`/`close`, write a Python script
and run it inside the container as shown in the subtask gotcha above. This
applies to: batch status changes, bulk ticket creation, inspecting parent
attributes, relationship wiring, etc.
### Pattern: before you act
- Always `redmine show <id>` before updating — confirm current status,
assignee, and existing notes so you don't clobber context.
- Never delete or close a ticket unless the user explicitly asks.
- When in doubt about project/tracker/priority, inherit from the parent or
ask the user.
## Git workflow
### Atomic commits
Each commit is **one logical change** — one feature, one fix, one doc update.
If you're tempted to write "and also..." in a commit message, that's a sign
to split it into two commits. Stage precisely (`git add <file>`, not
`git add -A`) so unrelated changes don't get bundled.
### Conventional commit messages
Use the [Conventional Commits](https://www.conventionalcommits.org/) format:
```
<type>(<optional scope>): <imperative subject>
<optional body — why, not what>
```
Types used in this repo: `feat`, `fix`, `docs`, `refactor`, `chore`, `style`.
Rules:
- Subject line **under 72 chars**, lowercase, imperative mood.
- No period at end of subject.
- Body wrapped at 72 chars, explains **why** the change exists.
### Commit cadence
- **Commit early and often.** Don't accumulate a pile of unrelated changes.
- Commit **without asking** — if you made a coherent change, commit it.
- Every task (ticket review, split, script addition, doc update) ends with
the relevant files committed.
- Run `git status` before committing to stage only what belongs together.
@@ -15,19 +15,21 @@ wrapper.
## Configuration
Copy the env template and fill in real values (the `.env` is gitignored):
Credentials live in the **centralized credential store** at
`~/.creds/redmine.env` (see `tooling-cli/KNELCredsManager`). It holds:
```bash
cp .env.example .env
# then edit .env:
# REDMINE_URL=https://your-redmine.example.com
# REDMINE_API_KEY=abc123...
```
REDMINE_URL=https://your-redmine.example.com
REDMINE_API_KEY=abc123...
```
Use `.env.example` in this directory as a template if you need to create one.
Permissions: directory `700`, the env file `600` (owner read/write only).
## Build
```bash
docker build -t kneldevstack-aimiddleware-redmine-cli .
docker build -t kneldevstack-aimiddleware-redmine-cli:latest .
```
A prebuilt image is also available in the Gitea registry:
@@ -38,21 +40,25 @@ git.knownelement.com/reachableceo/redmine-cli:latest
## Usage
### Via the repo wrapper (loads `.env` automatically)
Invoke the container directly with `docker run`. Pass credentials from the
centralized store via `--env-file`:
```bash
bin/redmine whoami
bin/redmine list --assigned-to-me
bin/redmine show 123
bin/redmine close 123 --notes "work complete"
docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest whoami
docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest list --assigned-to-me
docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest show 123
docker run --rm --env-file ~/.creds/redmine.env \
git.knownelement.com/reachableceo/redmine-cli:latest close 123 --notes "work complete"
```
### Via docker directly (from the repo root)
```bash
docker run --rm --env-file .env kneldevstack-aimiddleware-redmine-cli whoami
docker run --rm --env-file .env kneldevstack-aimiddleware-redmine-cli list --assigned-to-me
```
To use a locally built image instead of the registry image, either set
`REDMINE_CLI_IMAGE` or swap the image tag for `kneldevstack-aimiddleware-redmine-cli:latest`.
## Commands
@@ -94,4 +100,3 @@ docker run --rm --env-file .env kneldevstack-aimiddleware-redmine-cli list --ass
| ----------------- | -------- | ------------------------------------ |
| `REDMINE_URL` | yes | Base URL of the Redmine instance. |
| `REDMINE_API_KEY` | yes | API key of the acting user. |
| `REDMINE_CLI_IMAGE` | no | Override the image tag used by `bin/redmine` (default `git.knownelement.com/reachableceo/redmine-cli:latest`). |