/aitask-qa

Run QA analysis on any task — discover tests, run them, identify gaps, and create follow-up test tasks. This skill works with both active and archived tasks, making it useful for reviewing test coverage at any point in the task lifecycle.

Usage:

/aitask-qa              # Interactive: select from recently archived tasks
/aitask-qa 42           # Analyze a specific parent task
/aitask-qa 16_2         # Analyze a specific child task

Note: Must be run from the project root directory. See Skills overview for details.

This skill is read-only — it never modifies task status or claims ownership. It analyzes changes, runs tests, and optionally creates follow-up tasks for additional test coverage.

Step-by-Step

1. Profile selection — Loads an execution profile to pre-answer QA questions. See Execution Profiles for the profile schema
2. Task selection — Select a task by ID or interactively from recently archived/active tasks. Supports both parent and child tasks
3. QA tier selection — Choose analysis depth: Quick, Standard, or Exhaustive (see Three Tiers below)
4. Change analysis [Standard, Exhaustive] — Detects commits associated with the task, categorizes changed files into source code, test files, and config/docs
5. Test discovery [Standard, Exhaustive] — Scans for existing tests (bash test_*.sh, python test_*.py, TypeScript *.spec.ts, etc.), maps source files to test files, and identifies coverage gaps
6. Test execution [All tiers] — Runs discovered tests using configured commands or auto-detection, presents PASS/FAIL summary. Includes a health score [Standard, Exhaustive] and verification gate [Exhaustive]
7. Test plan proposal [Standard, Exhaustive] — Generates test proposals (unit, integration, edge cases) with regression test hints for bug-fix tasks. For Exhaustive tier, includes edge case brainstorming (concurrency, resource exhaustion, platform quirks, malformed input)
8. Follow-up task creation (optional) — Creates a follow-up test task as a sibling (for child tasks) or standalone task, pre-filled with the test plan

Three Tiers

The tier system controls analysis depth, letting you balance thoroughness against time:

Key Capabilities

• Works with archived tasks — Analyze any completed task, not just active ones. Select interactively from recently archived tasks or pass a task ID directly
• Commit-aware analysis — Detects commits using the (t<N>) pattern in commit messages, then categorizes changes to focus test proposals on modified code
• Auto-detection — When test_command is not configured, the skill auto-detects test files matching common patterns (tests/test_*.sh, test_*.py, *.spec.ts)
• Tiered depth — Quick tier for fast smoke tests, Standard for regular gap analysis, Exhaustive for critical-path verification with evidence tables
• Regression hints — For bug-fix tasks (issue_type: bug), generates specific regression test suggestions following the red-green verification cycle
• Follow-up task integration — Creates well-structured follow-up test tasks using the standard task creation pipeline, with proper dependencies and labels

Configuration

Profile Keys

These profile keys control /aitask-qa behavior when set in a profile YAML file (aitasks/metadata/profiles/):

Project Config Keys

These keys in aitasks/metadata/project_config.yaml configure test and lint commands:

See Build Verification for details on test_command, lint_command, and verify_build.

Workflows

For a workflow guide covering QA analysis patterns and when to use each tier, see QA and Testing.