agentleFS
roee-hershroee-hersh/toolpass/CLAUDE.md

toolpass

CLAUDE.md4 starsChanged 44 days agoApache-2.0View on GitHub

Do not stop at a pushed branch. A merge into main ships in the next daily release (03:00 UTC), so merge only complete, working changes. 1. Update the documentation for what changed, every time: the lead example and the tables in README.md and toolpass-py/README.md when the way a tool is written changes, the guide and reference pages under docs/ that describe the changed behaviour, docs/README.md when a page is added, and docs/development/testing.md when a test directory is added. A page…

  • Installs packages

The file

Show its rules and text

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

14 rules. About 0 are common, 1 is distinctive, and 13 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈1

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 13

    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
  1. Distinctive:

    Do not stop at a pushed branch.

    line 5
  2. Very specific:

    Update the documentation for what changed, every time: the lead example and the tables in README.md and toolpass-py/README.md when the way …

    line 8
  3. Very specific:

    Run /code-review main high on the diff and fix every finding.

    line 14
  4. Very specific:

    Run /security-review and fix every finding.

    line 15
  5. Very specific:

    Run the local checks, after the review fixes, and get them clean (in toolpass-py/, with pip install -e ".[crypto]" pytest pytest-timeout …

    line 16
  6. Very specific:

    when an integration or tests/harness changed: fetch the API descriptions with test/specs/fetch.sh /tmp/specs, then …

    line 20
  7. Very specific:

    when a framework adapter (src/toolpass/<framework>.py) changed: install its extra (pip install -e ".[<extra>]") and run python -m pytest …

    line 22
  8. Very specific:

    when the vault integration changed: TOOLPASS_REAL=1 python -m pytest tests/real (needs docker)

    line 24
  9. Very specific:

    when packaging (pyproject.toml) changed: build the wheel (python -m build -o /tmp/dist .), install it into a fresh venv, then run tests …

    line 25
  10. Very specific:

    when toolpass-ts changed: cd toolpass-ts && npm ci && npm test, and run the examples/agent-ts tests, which exercise it

    line 27
  11. Very specific:

    when examples/agent-ts changed: cd examples/agent-ts && npm ci && npm test

    line 29
  12. Very specific:

    when any Markdown changed: python3 test/docs/linkcheck.py

    line 30
  13. Very specific:

    when the Dockerfile changed: docker build .

    line 32
  14. Very specific:

    Wait for CI on the pull request to pass, then squash-merge with the PR title and (#N) suffix, as the history on main does.

    line 35

≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.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
CLAUDE.md

Working in this repository

Finishing a change

Do not stop at a pushed branch. A merge into main ships in the next daily release (03:00 UTC), so merge only complete, working changes.

  1. Update the documentation for what changed, every time: the lead example and the tables in README.md and toolpass-py/README.md when the way a tool is written changes, the guide and reference pages under docs/ that describe the changed behaviour, docs/README.md when a page is added, and docs/development/testing.md when a test directory is added. A page that shows the old way of doing something is a bug in the pull request, not a follow-up. Run python3 test/docs/linkcheck.py afterwards.
  2. Run /code-review main high on the diff and fix every finding.
  3. Run /security-review and fix every finding.
  4. Run the local checks, after the review fixes, and get them clean (in toolpass-py/, with pip install -e ".[crypto]" pytest pytest-timeout hypothesis ruff mypy types-PyYAML build):
    • ruff format --check src tests examples, ruff check src tests examples, mypy --strict src/toolpass, python -m pytest -q
    • when an integration or tests/harness changed: fetch the API descriptions with test/specs/fetch.sh /tmp/specs, then TOOLPASS_SPECS_DIR=/tmp/specs python -m pytest -q tests/integrations tests/harness_tests tests/contract
    • when a framework adapter (src/toolpass/<framework>.py) changed: install its extra (pip install -e ".[<extra>]") and run python -m pytest tests/frameworks/test_<framework>.py
    • when the vault integration changed: TOOLPASS_REAL=1 python -m pytest tests/real (needs docker)
    • when packaging (pyproject.toml) changed: build the wheel (python -m build -o /tmp/dist .), install it into a fresh venv, then run tests with that venv's Python from outside the repository
    • when toolpass-ts changed: cd toolpass-ts && npm ci && npm test, and run the examples/agent-ts tests, which exercise it
    • when examples/agent-ts changed: cd examples/agent-ts && npm ci && npm test
    • when any Markdown changed: python3 test/docs/linkcheck.py
    • when deploy/helm changed: helm lint deploy/helm/toolpass --strict --set apiKey.value=x
    • when the Dockerfile changed: docker build . and the smoke test in the docker CI job
  5. Open a pull request using .github/pull_request_template.md. Mention the issue it closes when there is one.
  6. Wait for CI on the pull request to pass, then squash-merge with the PR title and (#N) suffix, as the history on main does. If CI is red, fix and push; never merge over a failure, and never skip or disable a test to get green. A failure in code the change does not touch (a vendor spec, Prism or the argocd CLI failing to download, for example) may be re-run once; if it fails again, say so and stop.

The maintainer asked for this flow. Ask first only for changes that are destructive or clearly outside what was requested.

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.