agentleFS
JoseVelazcoHJoseVelazcoH/python-skills/AGENTS.md

python-skills

AGENTS.md0 starsChanged 5 months agoApache-2.0View on GitHub

Battle-tested Python skills, distributed as a Claude Code plugin. Built from custom engineering conventions. Blocks commits whose staged Python violates the skills, via Claude headless review. Note: this is an LLM review: non-deterministic and adds latency/token cost per commit. It catches judgment-level rules (cohesion, SOLID, hardcoded values) a linter cannot.

The file

Files
AGENTS.md

python-skills plugin

Battle-tested Python skills, distributed as a Claude Code plugin. Built from custom engineering conventions.

Skills

SkillTriggerCovers
python-clean-codeclean code, refactor, naming, comments, data classes, code smellsLine/function readability: naming, comments, classes-vs-functions, data types, smells, DRY/KISS/YAGNI
python-design-principlesSOLID, dependency inversion, coupling, strategy patternClass/module structure: SOLID, coupling/cohesion, Strategy
python-testing-tddwrite tests, pytest, TDD, fixtures, coverageRed-Green-Refactor + quality pytest patterns
python-code-reviewcode review, review PR, pre-mergeReview Pyramid, priority-first review
python-anti-patternsimport order, file too long, hardcoded values, module constantsCustom module-organization rules

Install in another repo

/plugin marketplace add <git-url-or-local-path-to-this-repo>
/plugin install python-skills@clean-python

Commands

  • /python-skills:review-rules: reviews a Python diff (from stdin) against the skill rules and emits VIOLATION:/NO_VIOLATIONS. Defined in commands/review-rules.md.

Pre-commit harness (per target project)

Blocks commits whose staged Python violates the skills, via Claude headless review.

  1. Install the plugin in the target project (above). Requires claude and jq on PATH.
  2. Copy assets/pre-commit.sh → .git/hooks/pre-commit in that project and chmod +x it.
  3. On each commit it pipes the staged diff to /python-skills:review-rules; violations block the commit. Bypass once with git commit --no-verify.

Note: this is an LLM review: non-deterministic and adds latency/token cost per commit. It catches judgment-level rules (cohesion, SOLID, hardcoded values) a linter cannot.

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.