Skip to content

TsukiNi22/ai-utils / skills/routers/python

Router for every Python task of the user - looks at the request and the project, then loads the right Python skills (python-project, python-class, python-style, python-comments, tests, ci-cd, benchmark). Invoke manually with /python.

Python router

<name> (a skill) = the folder of that skill: ~/.claude/skills/<name> (setup.sh) or ${CLAUDE_SKILL_DIR}/../../*/<name> (plugin of the marketplace).

Request given with /python: $ARGUMENTS

Do not answer from this file: decide which skills apply, load them with the Skill tool (never the other routers: they are manual-only), then do the task following them. The request is the text given with /python (or, if empty, the latest request of the user; if there is none, inspect the project and propose what can be done).

1. Look at the context (quickly, in parallel)

  • pyproject.toml (package), src/main.py + src/app.py (MAGIC application), a lone script, requirements.txt, tests/, .venv/. Empty folder = new project.
  • GitHub repository? git remote -v (github.com -> CI/CD through ci-cd).
  • Files the request is about (open them before deciding).

2. Pick the skills (several can apply, load all of them before writing)

SituationSkills
new project / package / tool / application / "setup"python-project (it loads the others it needs, ci-cd on GitHub)
new class, abstract class, tool function, module, exception, constants, test filepython-class (+ python-style, python-comments)
write / edit / refactor / fix Python codepython-style + python-comments
comments / docstrings onlypython-comments
review Python code against the user's stylepython-style + python-comments (xstyle --rtk -c PY,G first)
unit tests (pytest): setup, write, runtests
slow code, profiling, benchmarkbenchmark (py-spy / cProfile part)
CI / CD (tests, style, PyPI, release)ci-cd
commit / docs asked at the endgit-conventions / readme-style

Rules: python-class never writes function logic; the local style of an existing file wins.

3. Say it

Start the answer with one line: which skills were loaded and why (ex: "Skills : python-class, python-style, python-comments (nouvelle classe dans une application MAGIC)").

finishing-a-development-branchobra/superpowersUse when implementation is complete, all tests pass, and you need to decide how to integrate the workStars: 318kUpdated Claude CodeCodexUnverifiedInstall
systematic-debuggingobra/superpowersUse when encountering any bug, test failure, or unexpected behavior, before proposing fixesStars: 318kUpdated Claude CodeCodexUnverifiedInstall
test-driven-developmentobra/superpowersUse when implementing any feature or bugfix, before writing implementation codeStars: 318kUpdated Claude CodeCodexUnverifiedInstall
skill-creatoranthropics/skillsCreate new skills, modify and improve existing skills, and measure skill performance. Use when users want to create a skill from scratch, edit, or optimize an existing skill, run evals to test a skill, benchmark skill performance with variance analysis, or optimize a skill's description for better triggering accuracy.Stars: 196kUpdated Claude CodeCodexVerifiedInstall