qa-coordinator

QA Coordinator — Next-Action Engine

Determine the most helpful next action for the user based on current QA state.

When This Skill Is Invoked

Called at the end of every /qa:* command to append a "Suggested next step" line.

Decision Logic

After /qa:init

• If code exists and test types selected: "Setup complete. Run /qa:scenarios to generate user scenarios, or /test to run your tests."
• If code exists but no tests: "Setup complete. Use the test-writer skills to create your first tests, or /qa:scenarios to plan what to test."
• If no code: "Base setup complete. When you have code, run /qa:sync to select test types and generate artifacts."

After /qa:sync

• If config changed: "Configuration and artifacts updated. Run /qa:analyze to check test health."
• If only artifacts regenerated: "Artifacts updated. Run /qa:analyze to check test health against the new configuration."
• If everything current: "All artifacts are up to date. Run /qa:analyze for a deeper test health check."

After /qa:analyze

• If gaps found: "Found N coverage gaps. Highest priority: [module]. Use the test-writer skill to add tests."
• If quality issues: "Found N test quality issues. Run /test:review for detailed recommendations."
• If healthy: "Test health looks good. Run /test:coverage to verify coverage thresholds."

After /qa:scenarios

• If generated from spec: "Scenarios added to [spec file]. Use the e2e-test-writer skill to create e2e tests from these scenarios."
• If generated standalone: "Scenarios saved to .claude/scenarios/. Use the e2e-test-writer skill to create e2e tests from these scenarios, or run /qa:scenarios --spec <path> to attach them to a spec."

Output Format

Always output exactly one suggestion line at the end of the command output:

---
**Suggested next step:** Run `/qa:sync` to update 3 outdated artifacts.