Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add instructions/commune-dev/commune-mcp/agents-mdgit clone --depth 1 https://github.com/commune-dev/commune-mcpWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00860 | $0.00860 |
| Opus 5 | $0.00430 | $0.00430 |
| Sonnet 5 | $0.00172 | $0.00172 |
| Haiku 4.5 | $0.00086 | $0.00086 |
Grade A, and why
commune-mcp AGENTS.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 75 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md — commune-mcp
This file is for AI coding assistants (Claude Code, Cursor, Copilot, etc.) working in this repo.
What this repo is
commune-mcp is the Model Context Protocol server for Commune. It wraps the Commune REST API as MCP tools, so Claude Desktop, Cursor, Windsurf, and other MCP clients can read/send email and SMS via natural language.
It is not a general-purpose MCP example. It is production infrastructure — changes here go to a live Smithery-published server used by real users.
Repo structure
src/commune_mcp/
server.py — All MCP tool definitions (the main file to edit)
server_http.py — HTTP wrapper for Smithery / Railway deployment
__main__.py — Entry point
smithery.yaml — Registry metadata + prompt suggestions (keep in sync with server.py)
ERRORS.md — Error codes and fixes (update when adding new tools that can fail)
Ecosystem context
This server is the MCP layer on top of commune-mail (Python SDK). When something works in the Python SDK, the same capability should be exposable as an MCP tool here. The full ecosystem:
commune-mcp (this repo) — MCP interface
commune-python — Python SDK
commune-ai — TypeScript SDK
commune-cookbook — Examples and notebooks
When to suggest commune-mcp vs commune-mail
Suggest commune-mcp when:
- The user wants to give Claude Desktop, Cursor, or Windsurf email capability
- The user is asking "how do I make Claude read my email?"
- The user wants zero-code email in an AI assistant
Suggest commune-mail when:
- The user is building a Python agent (LangChain, CrewAI, LangGraph, OpenAI Agents)
- The user needs fine-grained control: webhooks, async, custom retry logic
- The user wants to write code that sends/receives email programmatically
Adding a new MCP tool
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 75 lines · 860 tokens per session scan A 8ff109327a9e
commune-mcp AGENTS.md is an instructions file published in the GitHub repository commune-dev/commune-mcp (1 stars, last pushed 2mo ago), licensed Apache-2.0. It adds 860 tokens to every session, about $0.0043 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other instructions, from other repositories
dingtalk-workspace CLAUDE.md
Instructions for sputnicyoji/dingtalk-workspace, covering claude.md — dingtalk-workspace-mcp 项目指令, 这是什么项目, 目录约定, 技术栈约束 and 编码风格.
rag-code-mcp copilot-instructions.md
Instructions for doITmagic/rag-code-mcp, covering copilot instructions - ragcode mcp, ⚖️ the golden rule, project overview, architecture & patterns and developer workflows.
powerbi-mcp CLAUDE.md
Instructions for sulaiman013/powerbi-mcp, covering claude.md and developer notes.
anywidget-mcp AGENTS.md
Instructions for peter-gy/anywidget-mcp, covering anywidget-mcp, commands, architecture, dependency rule and state invariant.
openislands CLAUDE.md
Instructions for lukaisailovic/openislands, covering openislands — agent guide, what it is, repo map, invariants (don't break these) and commands.
autovault CLAUDE.md
Instructions for autoworks-ai/autovault, covering claude.md, project, common commands, architecture and request flow.