# Local QA

> Run local QA including formatting, linting, type-checking, static site build, and workflow checks for the repository. Use whenever any file has been updated.

- Skill: `dceoy/local-qa` (Agent Skill, multi-file: 2 files)
- Install (CLI): `npx skillmds@latest add dceoy/local-qa`
- Raw SKILL.md: https://api.skillmd.com/api/skills/dceoy/local-qa/raw
- Safety review: pending
- Works with: Claude Code, Claude.ai, OpenAI Codex
- Category: Productivity, Coding & Dev Tools
- Author: dceoy (https://skillmd.com/u/dceoy)
- Updated: 2026-09-17
- Page: https://skillmd.com/skills/dceoy/local-qa

---


# Local QA (format, lint, type-check, and build)

Run the local QA script for this repository:

```bash
.agents/skills/local-qa/scripts/qa.sh
```

## Procedure

1. Execute the script exactly as shown above when this skill is triggered.
2. Capture and summarize key output: success/failure, major warnings, and any files modified.
3. If the script fails due to missing tooling (`command not found`, missing executable, or equivalent), install the missing tool(s) and rerun `.agents/skills/local-qa/scripts/qa.sh`.
4. Install tools using this order of preference:
   1. Use the project's package manager when applicable (`uv` for Python, package manager scripts/dependencies for Node.js).
   2. Use a system package manager (`brew` on macOS, `apt` on Debian/Ubuntu) when project-local install is not applicable.
   3. Use language-specific installers as fallback (`pipx`/`pip`, `npm`, `go install`, etc.).
5. If multiple tools are missing, repeat install -> rerun until QA completes or you hit a blocker.
6. If installation fails or requires unavailable privileges, report what was attempted, the exact failure, and stop.
7. Do not run unrelated commands; only run commands needed for QA and missing-tool installation.