From autonomous-sdlc
Runs Python verification pipeline: ruff format/check --fix, mypy type checking, pytest tests in fast-fail sequence. Use before commits/merges/PRs or on triggers like 'run all checks', 'verify code'.
npx claudepluginhub joshuaoliphant/claude-plugins --plugin autonomous-sdlcThis skill uses the workspace's default tool permissions.
Run automated quality gates (format, lint, type check, test) as deterministic verification instead of manual approval. Verification replaces permission prompts — if checks pass, proceed; if they fail, fix and retry.
Runs code verification phases including build, type check, lint, tests, security scans, and diff review for Python and Node.js projects before PRs.
Runs lint, type-check, tests, and build checks for Node.js/TS, Python, Rust, Go, Java projects to verify code health after changes.
Enforces five-step verification (identify/run/read/verify/claim) of commands like npm test/build/lint before claiming code complete, fixed, or passing. Builds trust via evidence.
Share bugs, ideas, or general feedback.
Run automated quality gates (format, lint, type check, test) as deterministic verification instead of manual approval. Verification replaces permission prompts — if checks pass, proceed; if they fail, fix and retry.
Core Principle — Asymmetry of Verification: Many tasks are easier to verify than to solve. Software development is highly verifiable through tests, linters, type checkers, and build systems.
| Language | Format | Lint | Types | Test |
|---|---|---|---|---|
| Python | ruff format | ruff check | mypy | pytest |
| TypeScript | pnpm format | pnpm lint | tsc | vitest |
| Go | go fmt | golangci-lint | go build | go test |
Run checks in this order — stop at first failure:
uv run ruff format .uv run ruff check . --fixuv run mypy src/uv run pytest tests/ -x --tb=shortuv run pytest tests/ --cov=src/uv run bandit -r src/ --severity-level high| Failure | Fix |
|---|---|
| Lint: unused import | Remove the import |
| Type: missing return | Add return type annotation |
| Test: assertion failed | Fix logic or update test |
| Format: style violation | Auto-fix handles this |
python ${CLAUDE_PLUGIN_ROOT}/scripts/feedback_manager.py autonomous-sdlc show-feedback
Apply relevant feedback: verification, general.
For Python projects:
uv run ruff format . && uv run ruff check . --fix && uv run mypy src/ && uv run pytest tests/ -x --tb=short
In autonomous workflows:
# Quick check during development
uv run pytest tests/test_specific.py -x
# Full verification before closing Bead
uv run ruff format . && uv run ruff check . --fix && uv run mypy src/ && uv run pytest tests/
A green or red verification result. Green means proceed. Red means fix and retry. No human approval needed — the pipeline is the gate.
For projects that want a single entry point:
#!/bin/bash
# scripts/verify.sh
set -e
echo "=== Formatting ===" && uv run ruff format .
echo "=== Linting ===" && uv run ruff check . --fix
echo "=== Type Checking ===" && uv run mypy src/
echo "=== Tests ===" && uv run pytest tests/ -x --tb=short
echo "=== All checks passed ==="