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/oddrationale/groupme-mcp-server/agents-mdgit clone --depth 1 https://github.com/oddrationale/groupme-mcp-serverWhat 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 | $0.02484 | $0.02484 |
| Opus 5 | $0.01242 | $0.01242 |
| Sonnet 5 | $0.00497 | $0.00497 |
| Haiku 4.5 | $0.00248 | $0.00248 |
Grade A, and why
groupme-mcp-server 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 — 225 lines — stays where its author put it; the contents beside it link to each section on GitHub.
AGENTS.md
Instructions for AI coding agents (Claude Code, Copilot, Cursor, …) working in this repository. Humans: see CONTRIBUTING.md.
What this is
groupme-mcp-server is a FastMCP server exposing the
GroupMe API v3 over MCP. It is deployed to
Prefect Horizon.
Current state: read, agentic, and write tools implemented
(list_conversations, read_messages, get_conversation_context,
search_messages, get_highlights, send_message, react_to_message);
image upload (arbitrary-URL images) is not yet.
Toolchain — use these, not the alternatives
| Concern | Tool | Do not use |
|---|---|---|
| Packaging | uv (src layout) |
pip, poetry, pdm, hatch |
| Lint + format | ruff |
black, flake8, isort, pylint |
| Type check | ty |
mypy, pyright |
| Tests | pytest + pytest-cov |
unittest, nose |
| Git hooks | lefthook |
pre-commit |
Every command runs through uv: uv run <cmd>, uv add <pkg>,
uv add --dev <pkg>. Never call pip or edit uv.lock by hand.
Exception: on Windows ARM64 the lefthook PyPI wrapper cannot find its binary
(upstream arch-detection bug), so lefthook is installed natively via scoop and
invoked as bare lefthook, not uv run lefthook.
Before you claim you are done
uv run ruff format . && uv run ruff check . && uv run ty check && uv run pytest
All four must pass. There is no partial credit.
Integration tests
tests/integration/ holds tests marked integration that talk to real
services: read-only smoke tests against api.groupme.com, auth checks
against the deployed server at https://groupme.fastmcp.app/mcp, and a
stdio end-to-end test that spawns uv run groupme-mcp-server as a
subprocess. They are deselected by default (-m "not integration" in
addopts), so the ordinary gate, CI, and the 100% coverage requirement
never touch them. Run them with:
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 · 225 lines · 2,484 tokens per session scan A fde96fe9a459
groupme-mcp-server AGENTS.md is an instructions file published in the GitHub repository oddrationale/groupme-mcp-server (0 stars, last pushed 4d ago), licensed MIT. It adds 2,484 tokens to every session, about $0.0124 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
swag-mcp CLAUDE.md
Instructions for jmagar/swag-mcp, covering swag mcp - claude memory reference, project overview, core architecture, key components and common development commands.
sbb-opendata-mcp CLAUDE.md
Instructions for malkreide/sbb-opendata-mcp, covering claude.md, teil 1 — konventionen (portfolio-weit), vor der arbeit, tests and wenn etwas rot ist.
whatsapp-mcp-server CLAUDE.md
Instructions for HalemoGPA/whatsapp-mcp-server, covering working on this repo, the one idea, rules that follow from it, the box has a hard resource budget - stay inside it and repo-specific traps.
microsoft-planner-mcp python.instructions.md
Use when writing or modifying Python source code. Covers module layout, async patterns, imports, and framework conventions for FastMCP + Microsoft Graph SDK.
microsoft-planner-mcp testing.instructions.md
Use when writing or modifying tests. Covers pytest-asyncio auto mode, parameterized test conventions, mock patterns, and test file structure for this project.
todoist-mcp-server CLAUDE.md
Claude Code instructions for stevesimpson418/todoist-mcp-server, covering claude.md — project intelligence, project overview, tech stack, project structure and key commands.