MorgenMCP
AGENTS.md3 starsChanged 4 months agoApache-2.0View on GitHub
MorgenMCP is a Python MCP server that exposes the Morgen calendar, task, and tag APIs through FastMCP. The package source is in morgenmcp/; tests are in tests/. - Use Python 3.12 or newer (development pins .python-version) and manage dependencies with uv. 3.12 is a hard floor — models.py uses PEP 695 generic syntax — so keep annotations valid there: every module imports from __future__ import annotations, and CI tests the floor as well as 3.14. - Install development dependencies with…
- Reads credentials
Where files are headingNew AGENTS.md files have more rules about keeping secrets out of the code, and fewer about code style.
Of the AGENTS.md files we read, the 32,671 first seen in the last six months, against the older ones: the kind of rule that gained most share, and the kind that lost most. About files like this in general, not this one, and an estimate, since a kind is a pattern in the words.
The file
Show its rules and text
We split this AGENTS.md into the rules it gives an agent
13 rules. About 1 is common, 7 are distinctive, and 5 are very specific to this project.
≈1
Common · estimate
Many files like this have a rule of this kind.
≈7
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- could be automatic: a setting could enforce it
- Very specific: line 11
Use Python 3.12 or newer (development pins .python-version) and manage dependencies with uv.
- Very specific: line 17
Run the default test suite with uv run pytest.
- Distinctive: line 20
Before handing off Python changes, run:
- Distinctive: line 28
Follow the existing async style for API, tool, and resource code.
- Distinctive: line 29
Keep user-facing MCP inputs and outputs validated with the existing Pydantic models and validator helpers.
- Common: line 33
Do not log API keys, raw IDs that can reveal account information, or other secrets.
- Very specific: line 35
Add or update focused tests alongside behavior changes.
- Distinctive: line 40
Do not modify generated lockfiles or dependency versions unless the task requires it.
- Very specific: line 44
Keep .env and credentials local.
- Very specific: line 48
Read and follow AI_POLICY.md.
- Distinctive: line 51
Verify claims against this repository (run the command, check output) before reporting.
- Distinctive: line 52
Ensure a human understands and can explain all submitted changes.
- Distinctive: line 67
If your change duplicates work already proposed in another open PR, say so rather than leaving it for a reviewer to discover.
≈ “Common” is an estimate: at least 1 in 5 of the 41,581 AGENTS.md files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
MorgenMCP Contributor Guide
Project overview
MorgenMCP is a Python MCP server that exposes the Morgen calendar, task, and
tag APIs through FastMCP. The package source is in morgenmcp/; tests are in
tests/.
Environment and commands
- Use Python 3.12 or newer (development pins
.python-version) and manage dependencies withuv. 3.12 is a hard floor —models.pyuses PEP 695 generic syntax — so keep annotations valid there: every module importsfrom __future__ import annotations, and CI tests the floor as well as 3.14. - Install development dependencies with
uv sync --all-extras. - Set up git hooks once with
pre-commit install. - Run the default test suite with
uv run pytest. Integration tests are excluded by default; run them explicitly withuv run pytest -m integrationonly whenMORGEN_API_KEYis configured and real API calls are intended. - Before handing off Python changes, run:
uv run ruff check .uv run ruff format .uv run pyright morgenmcp/- relevant tests (or the full non-integration suite when practical)
Code conventions
- Follow the existing async style for API, tool, and resource code.
- Keep user-facing MCP inputs and outputs validated with the existing Pydantic models and validator helpers.
- Preserve virtual-ID behavior in
morgenmcp/tools/id_registry.py; callers should not expose raw Morgen IDs. - Do not log API keys, raw IDs that can reveal account information, or other secrets. Keep persistent ID storage owner-only.
- Add or update focused tests alongside behavior changes. Mock HTTP requests
with the existing
respxpatterns; do not let normal tests call the live API.
Scope and safety
- Do not modify generated lockfiles or dependency versions unless the task requires it.
- Preserve public MCP tool names and parameter compatibility unless a breaking change is explicitly requested.
- Keep
.envand credentials local; never add them to version control.
AI usage
- Read and follow AI_POLICY.md.
- Disclose all AI assistance in the PR description (tool and extent).
- Mark AI-assisted commits with
Assisted-by: LLM (<tool>, <model>); do not useCo-Authored-By. - Verify claims against this repository (run the command, check output) before reporting.
- Ensure a human understands and can explain all submitted changes.
Working alongside other agents
Multiple agents or harnesses may work on this repo at the same time, each in
its own branch or worktree off main. morgenmcp/client.py,
morgenmcp/server.py, and morgenmcp/tools/outputs.py are touched by nearly
every tool-adding change and are the most likely files to conflict between
concurrent branches — this has already happened in practice (two same-day PRs
both rewrote MorgenClient.list_tasks). Before opening a PR:
- Rebase onto the latest
main. - Skim the other open PRs' file lists for overlap with files you changed; note any expected conflict and how to resolve it (which side's fix must survive) directly in your PR description.
- If your change duplicates work already proposed in another open PR, say so rather than leaving it for a reviewer to discover.
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
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.

