quality-hooks
Scannednpx machina-cli add skill a5c-ai/babysitter/quality-hooks --openclawquality-hooks
You are quality-hooks -- the quality pipeline skill for Pilot Shell.
Overview
This skill defines the language-specific quality tool chains that run as PostToolUse hooks in Pilot Shell. It auto-detects the project language and applies the appropriate linter, formatter, and type checker with auto-fix support.
Tool Chains
Python
ruff check --fix . # Lint with auto-fix
ruff format . # Format
pyright . # Type check
TypeScript / JavaScript
npx eslint --fix . # Lint with auto-fix
npx prettier --write . # Format
npx tsc --noEmit # Type check
Go
golangci-lint run --fix # Lint with auto-fix
gofmt -w . # Format
go vet ./... # Type/static check
Auto-Fix Convergence Loop
When quality checks fail:
- Apply all auto-fixable issues
- Re-run quality checks
- Repeat up to 3 times
- Report remaining unfixable issues
Quality Scoring
Composite score = weighted average:
- Lint: 25%
- Format: 15%
- Typecheck: 25%
- Tests: 35%
Target: 85/100 (configurable via targetQuality)
Hook Integration Points
| Pilot Shell Hook | Quality Action |
|---|---|
| PostToolUse (Edit/Write) | Run file_checker on modified files |
| PostToolUse (file creation) | Run full pipeline on new files |
| Pre-merge | Run full pipeline on all changes |
Source
git clone https://github.com/a5c-ai/babysitter/blob/main/plugins/babysitter/skills/babysit/process/methodologies/pilot-shell/skills/quality-hooks/SKILL.mdView on GitHub Overview
quality-hooks is a language-specific quality pipeline for Pilot Shell. It auto-detects the project language and runs the appropriate linter, formatter, and type checker with auto-fix support, operating as PostToolUse hooks to keep code clean.
How This Skill Works
The tool auto-detects the project language and executes the corresponding chain: Python uses ruff check --fix, ruff format, and pyright; TypeScript/JavaScript uses npx eslint --fix, npx prettier --write, and npx tsc --noEmit; Go uses golangci-lint run --fix, gofmt -w, and go vet ./.... If quality checks fail, it enters an auto-fix convergence loop: apply fixes, re-run checks, repeat up to 3 times, then report remaining issues. It computes a composite score using weighted categories (Lint 25%, Format 15%, Typecheck 25%, Tests 35%), with targetQuality configurable (default 85).
When to Use It
- After editing Python, TypeScript/JavaScript, or Go files to catch issues early
- Before merging changes to ensure clean diffs
- When CI reports lint/format/typecheck failures in a PR
- During PostToolUse hooks to auto-fix and recheck modified files
- When adding new files to automatically apply quality checks from the start
Quick Start
- Step 1: Install/enable quality-hooks in Pilot Shell and ensure required tools (ruff, pyright, eslint, prettier, tsc, golangci-lint) are available
- Step 2: Configure targetQuality in your settings (default 85)
- Step 3: Make code changes; quality-hooks will auto-fix, re-check, and report remaining issues up to 3 iterations
Best Practices
- Enable auto-fix where safe to minimize manual edits
- Run the full pipeline on new or modified files before commits or merges
- Tune targetQuality to match your project's quality bar
- Review the final report to address unfixable issues
- Keep all tooling up to date (ruff, pyright, eslint, prettier, tsc, golangci-lint)
Example Use Cases
- A Python project auto-fixes lint errors with ruff and formats code with ruff format, then type-checks with pyright
- A TypeScript project runs eslint --fix, formats with prettier, and type-checks with tsc
- A Go project lints with golangci-lint --fix, formats with gofmt, and runs go vet
- A PR pre-merge triggers the full pipeline across all changes for consistency
- New files trigger a complete quality pass to establish a clean baseline from day one
