agentleFS
Sign inStart free
sandraschisandraschi/comms-mcp/CLAUDE.md

comms-mcp

CLAUDE.md1 starsChanged 12 days agoMITView on GitHub

Unified fleet comms gateway (Telegram / WhatsApp / Slack / Teams) behind the comms_ops portmanteau. Allowlist-gated send, sanitized inbound, 7-day retention, SQLite outbox. Backend :11205, console :11204, wa-sidecar :11208. - uv run python -m comms_mcp — stdio; MCP_PORT=11205 → HTTP daemon - just serve | test | lint | fix | fmt | types | ci | bootstrap | mcpb-pack - Console: web_sota/ (bun run dev, proxied to :11205 in dev, served from :11205 root via web_sota/dist in daemon mode)…

The file

Create a free account to mount
Files
CLAUDE.md

CLAUDE.md — comms-mcp

Unified fleet comms gateway (Telegram / WhatsApp / Slack / Teams) behind the comms_ops portmanteau. Allowlist-gated send, sanitized inbound, 7-day retention, SQLite outbox. Backend :11205, console :11204, wa-sidecar :11208.

Entry points

  • uv run python -m comms_mcp — stdio; MCP_PORT=11205 → HTTP daemon
  • just serve | test | lint | fix | fmt | types | ci | bootstrap | mcpb-pack
  • Console: web_sota/ (bun run dev, proxied to :11205 in dev, served from :11205 root via web_sota/dist in daemon mode)
  • NSSM: scripts/install-services.ps1 (elevated) — comms-mcp + comms-mcp-wa

Standards that bite here

  • TOOL_DESIGN_STANDARDS.md §4.2 dialogic returns (success + message), §7.1 error handling (_error_response + logger.exception), §9 MCP annotation hints — not custom keys.
  • STARLETTE_NO_PYDANTIC_STANDARD.md — the /mcp mount pattern (http_app(path="/") + mount("/mcp"), bare-path rewrite middleware, composed mcp_http.lifespan(app)); verify with a real fastmcp.Client("http://127.0.0.1:PORT/mcp"), never in-process ASGI.
  • Fleet ruff: T20 (no print in server code) enforced; per-file-ignores for CLI/build/test paths only.

Key files

  • src/comms_mcp/server.py — FastAPI + /mcp mount + REST
  • src/comms_mcp/__main__.py — dual transport + stdio→daemon probe/proxy
  • src/comms_mcp/mcp/tools/comms.py — the portmanteau; admin.py — shutdown
  • src/comms_mcp/adapters/*.py — one module per channel (allowlist + sanitize)
  • src/comms_mcp/outbox/sqlite.py — outbox + inbound store, TTL purge
  • skills/comms/SKILL.md — agent domain skill (also GET /api/skills)
  • docs/ONBOARDING.md — per-channel setup; docs/TROUBLESHOOTING.md — fixes

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

Check your setup, free