Claude Code skills distilled from the Software Quality knowledge base — the ways of finding out whether software works, across types, static analysis, tests, contracts, fuzzing, monitoring and mathematical proof. Each method carries its evidence, or a note where it is thin; practices that rest on repetition rather than data are named as folklore.
Warning
Experimental. The knowledge base was built from scratch, and these skills were then developed from it. Neither has had much tuning or testing against real projects yet, so expect rough edges — check the recommendations before acting on them.
/plugin marketplace add stereobooster/software-quality-skills
/plugin install quality-strategy@software-quality
/plugin install python-typing@software-quality
/reload-plugins
The plugins are independent — install either one on its own.
| Plugin | What it does |
|---|---|
quality-strategy |
Plans or audits a project's quality posture across the whole method space — types, static analysis, tests of every kind, contracts, fuzzing, monitoring, formal methods, code review, supply-chain. Use it when you'd otherwise ask "what tests should I write?"; it asks what uncertainty needs reducing and picks the method that reduces it. Produces a markdown report. |
python-typing |
Corrects how Python type annotations get written by default — typing the boundaries with pydantic rather than bypassing them, narrowing rather than casting, generating the types that tools can generate, and not laundering Any past a strict checker. For pyright and mypy. |
The skills are MIT (see LICENSE). The knowledge-base prose they draw on is
CC BY-SA 4.0.
The skills are authored in a private repository and mirrored here by CI on every change. Edits committed directly to this repository are overwritten by the next sync — please open an issue instead of a pull request.