readoc
CLAUDE.md0 starsChanged 7 days agoMITView on GitHub
Make sure to check all of the following and make sure they are up-to-date after making changes; 1. tool-specific documentation for tools you edited 2. skills for tools you edited 3. plugin.json 4. README.md 5. CLAUDE.md 6. tests/ — keep the pytest suite green and add coverage for behaviour you change Bump the plugin version on every commit. Patch version for small fixes, minor version for more substantial changes (new skill or tool). The bin/readoc, bin/readir, and bin/editdoc CLIs are…
In this repositoryOne other file it gives its agents.
- office-documentsSkill
The file
We split this CLAUDE.md into the rules it gives an agent
1 rule. About 0 are common, 1 is distinctive, and 0 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.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 1
Make sure to check all of the following and make sure they are up-to-date after making changes.
≈ “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.
Files
Make sure to check all of the following and make sure they are up-to-date after making changes;
- tool-specific documentation for tools you edited
- skills for tools you edited
- plugin.json
- README.md
- CLAUDE.md
- tests/ — keep the pytest suite green and add coverage for behaviour you change
Bump the plugin version on every commit. Patch version for small fixes, minor version for more substantial changes (new skill or tool).
Tests
The bin/readoc, bin/readir, and bin/editdoc CLIs are covered by a pytest
suite in tests/.
Dev tooling is managed by pyproject.toml (a dev-only virtual uv project,
package = false). Before changing any of these scripts, establish a green
baseline:
uv sync # first time / after dependency changes
uv run pytestLint and format with uv run ruff check . and uv run ruff format .. Tests
generate their own fixtures in a temp dir and run the CLIs as subprocesses; see
README.md (Development / Testing) for details.
hk.pkl (via hk, provisioned by mise.toml) runs ruff-check, ruff-format,
pytest, the audits (vulture dead-code, jscpd duplication), gitleaks secret
scanning, and a large-file guard on pre-commit; the same checks run in CI
(.github/workflows/ci.yml) on push/PR to master. After changing the dev
workflow, keep all three in sync. Run the whole suite locally with
hk run check --all.
This repo follows Mick's dev-env standard (dev-hooks:dev-env-setup), version
stamped via DEV_ENV_VERSION in mise.toml (currently v31). Tool versions
are pinned reproducibly in the committed mise.lock; bump them with
mise upgrade (commit the diff). uv enforces a 3-day dependency cooldown
([tool.uv] exclude-newer). Key dev tooling: hk + pkl (pre-commit), uv (Python
env, 3.12 in CI), ruff (lint/format), pytest, vulture (dead code), jscpd
(duplication, via npx), gitleaks (secrets). .gitleaks.toml allowlists
gitignored runtime paths so gitleaks dir's whole-tree scan stays green.
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.