Branch and commit conventions that keep an agent's work reviewable — one branch per task, main stays clean, nothing force-pushed.
日本語の概要は準備中です。原文の説明を表示しています。
Run the project's own tests, linter, type checker and build, and report what actually failed.
インストール方法を見るインストールする前に、エージェントに与えられる指示の中身を確認できます。
Run all quality checks: Tests → Lint → Types → Build
"Find bugs before they find users. Run QA early, run it often."
# Check what we're working with
[ -f "package.json" ] && echo "node"
[ -f "Cargo.toml" ] && echo "rust"
[ -f "pyproject.toml" ] || [ -f "requirements.txt" ] && echo "python"
[ -f "go.mod" ] && echo "go"
# Tests
npm test 2>&1 || yarn test 2>&1 || pnpm test 2>&1
# Lint
npm run lint 2>&1 || npx eslint . 2>&1
# Type check
npx tsc --noEmit 2>&1
# Build
npm run build 2>&1
# Tests
cargo test 2>&1
# Lint
cargo clippy -- -D warnings 2>&1
# Format check
cargo fmt --check 2>&1
# Build
cargo build 2>&1
# Tests
pytest 2>&1 || python -m pytest 2>&1
# Lint
ruff check . 2>&1 || flake8 . 2>&1
# Type check
mypy . 2>&1 || pyright . 2>&1
# Format check
black --check . 2>&1 || ruff format --check . 2>&1
# Tests
go test ./... 2>&1
# Lint
golangci-lint run 2>&1
# Build
go build ./... 2>&1
Format output as:
## QA Report
### Tests
✓ Passed: 42
✗ Failed: 2
⊘ Skipped: 1
Failed tests:
- test_user_auth: AssertionError at line 45
- test_token_refresh: Timeout
### Lint
✓ No issues
### Type Check
⚠ 3 warnings:
- src/auth.ts:12 - Type 'any' used
- src/utils.ts:34 - Implicit return type
### Build
✓ Build successful
### Summary
Status: FAILING
Must fix: 2 test failures
Should fix: 3 type warnings
Based on results:
If all pass:
QA passed! Ready for:
- /dev-security - Security audit
- /dev-rc - Release candidate prep
- /dev-finish - Close the cycle
If failures:
QA found issues.
Options:
1. Fix now (I'll help)
2. Fix manually, run /dev-qa again
3. Continue anyway (not recommended)
If "Fix now":
For a fast check (CI-style):
/dev-qa quick
Runs only:
Skips:
If active dev-cycle session exists:
## QA History
- 2026-07-15 14:30: FAILED (2 test failures)
- 2026-07-15 14:45: PASSED
| Check | Why |
|---|---|
| Tests | Does it work? |
| Lint | Is it clean? |
| Types | Is it safe? |
| Build | Does it compile? |
まだレビューはありません。使ってみた感想をお寄せください。
概要と使いどころ
Branch and commit conventions that keep an agent's work reviewable — one branch per task, main stays clean, nothing force-pushed.
日本語の概要は準備中です。原文の説明を表示しています。
Look up how Claude Code itself works — hooks, skills, subagents, MCP, plugins, settings — from the official docs, so answers come from the current spec instead of memory.
日本語の概要は準備中です。原文の説明を表示しています。
Design better agents, skills and hooks for Claude Code, checking the current spec instead of guessing at fields.
日本語の概要は準備中です。原文の説明を表示しています。
Run a feature end to end — scout the code, plan it, build it, and keep a session file so a closed terminal doesn't lose the thread.
日本語の概要は準備中です。原文の説明を表示しています。
Close out a session — summarise what changed, capture what was learned, and prepare the commit.
日本語の概要は準備中です。原文の説明を表示しています。
Hand over a spec and walk away — plans, builds, tests and audits on its own, stopping only at a real blocker.
日本語の概要は準備中です。原文の説明を表示しています。