Build, Test & Lint Autopilot

Build, Test & Lint Autopilot

Purpose

Provide Claude with a turnkey workflow for orchestrating the project's automated quality checks, including when to run full suites and how to interpret failures.

Activation cues

• Requests to run tests, lint, or end-to-end quality checks.
• Follow-up questions after CI failures.
• Planning sessions that need a recommended verification order.

Primary commands

Scenario	Command
Standard unit tests	./run_tests.sh
Integration focus	./run_tests.sh --integration
Coverage report	./run_tests_with_coverage.sh
UI/browser suite	./run_ui_tests.sh
Lint & format	./run_lint.sh or pre-commit run -a

Automation helper

Use the bundled script to run tests followed by linting in one step:

skills/build_test_lint_autopilot/scripts/run_quality_suite.sh [<run_tests.sh flags>]

The script executes from repo root, forwards any run_tests.sh options (e.g., --integration, --full), and stops on the first failure.

Decision guide

1. Small backend change → ./run_tests.sh
2. Firestore or external APIs → add --integration
3. Pre-merge confidence → ./run_tests_with_coverage.sh
4. Frontend tweaks → ./run_ui_tests.sh
5. Any code touch → ./run_lint.sh (or bundled script)

Failure triage tips

• Unit failures: Inspect stack traces in latest_ci_logs.txt or re-run specific modules via ./run_tests.sh path/to/test.py -k name.
• Integration issues: Confirm TEST_MODE environment and check service credentials in .env.
• Lint errors: Run pre-commit run -a for auto-fixes; review Ruff output for rule IDs.
• UI flakes: Re-run with ./run_ui_tests.sh --headed when available to capture screenshots.

Reporting expectations

• Summarize command(s) executed and whether they passed.
• Highlight failing modules and recommend next steps (e.g., enable verbose logging, update fixtures).
• Suggest follow-up commands such as pytest -k filters or rerunning coverage after fixes.