agentleFS
rodos4312rodos4312/Meu-Agent/pm/AGENTS.md

Meu-Agent / pm

AGENTS.md0 starsChanged 10 days agoMITView on GitHub

Applies on top of the root AGENTS.md (which carries the short form of the pinning rule). All dependencies carry upper bounds (litellm compromise #2796/#2810; Mini Shai-Hulud worm, May 2026). PyPI: >=floor,<next_major ("httpx>=0.28.1,<1"); pre-1.0: <0.(minor+2) (>=0.29,<0.32). Git URLs: 40-char commit SHA. GitHub Actions: SHA + # vN comment. CI-only Python requirements: ==exact. A bare >=X.Y.Z is rejected by CI and reviewers. After changing pyproject.toml, run hermes pm lock, re-source ./activate, and commit pyproject.toml with uv.lock. Reference: #2810 (bounds), #9801 (SHA pinning…

The file

We split this AGENTS.md into the rules it gives an agent

3 rules. About 0 are common, 0 are distinctive, and 3 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈0

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 3

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    All dependencies carry upper bounds (litellm compromise #2796/#2810.

    line 7
  2. Very specific:

    After changing pyproject.toml, run hermes pm lock, re-source ./activate, and commit pyproject.toml with uv.lock.

    line 11
  3. Very specific:

    Use pm.build_environment for fresh build outputs and pm.ensure_environment for isolated tool environments.

    line 17

≈ “Common” is an estimate: at least 1 in 5 of the 40,246 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.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
AGENTS.md

pm/ — dependencies and Hermes environments

Applies on top of the root AGENTS.md (which carries the short form of the pinning rule).

Dependency pinning policy

All dependencies carry upper bounds (litellm compromise #2796/#2810; Mini Shai-Hulud worm, May 2026). PyPI: >=floor,<next_major ("httpx>=0.28.1,<1"); pre-1.0: <0.(minor+2) (>=0.29,<0.32). Git URLs: 40-char commit SHA. GitHub Actions: SHA + # vN comment. CI-only Python requirements: ==exact. A bare >=X.Y.Z is rejected by CI and reviewers. After changing pyproject.toml, run hermes pm lock, re-source ./activate, and commit pyproject.toml with uv.lock. Reference: #2810 (bounds), #9801 (SHA pinning + audit CI).

PM owns Hermes Python dependency changes. Use pm.sync_venv(['extra'], explicit=True) for declared runtime extras, hermes pm install for setup/sync, and hermes pm repair for damaged dependencies. Do not mutate Hermes environments with raw pip or uv. Use pm.build_environment for fresh build outputs and pm.ensure_environment for isolated tool environments. Callers receive an interpreter or tool path, not uv. Nix's declarative uv2nix builds and unrelated user projects remain independently owned.

The [tool.uv] exclude-newer = "3 days" release-age check covers Hermes's own dependencies only, exact pins included, with one per-package exemption class: CA trust stores (certifi), whose revocation releases must reach the lock the day they ship (exclude-newer-package; nothing else goes there). It runs when Hermes LOCKS (hermes pm lock, CI's lock check), never on users' machines: every generated workspace drops the cutoff (pm/workspace.py::_without_release_age), so a mirror without upload times cannot break an install (#135589). Plugin resolution is unchanged by this: core's exact pins stay put, and a plugin may move a transitive core package within core's declared ranges (tests/pm/test_workspace_build_inputs.py::test_plugin_can_move_compatible_transitive_but_not_exact_requirement). Plugin python_dependencies follow the plugin's own policy. Teknium's rulings: "plugins dont have to abide by our 14 day rule … Only hermes' dependencies themselves have to"; dates are checked by us, not users; pinned deps need >3 days.

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.