Context
-
Project files: !find . -maxdepth 1 ( -name 'pyproject.toml' -o -name 'package.json' -o -name 'Cargo.toml' -o -name 'go.mod' )
-
Test directories: !find . -type d ( -name "test*" -o -name "tests" )
-
E2E setup: !find . -maxdepth 1 ( -name 'playwright.config.' -o -name 'cypress.config.' )
-
CI environment: !echo "CI=$CI GITHUB_ACTIONS=$GITHUB_ACTIONS"
Parameters
-
--coverage : Generate coverage report
-
--parallel : Force parallel execution
-
--report : Generate detailed HTML report
Your task
Delegate this task to the test-runner agent.
Use the Agent tool with subagent_type: test-runner to run the complete test suite. Pass all the context gathered above and specify All Tiers execution.
The test-runner agent should:
Run tests in pyramid order (fail fast):
-
Tier 1 - Unit tests first (fastest feedback)
-
Tier 2 - Integration tests (component interactions)
-
Tier 3 - E2E tests (full user flows)
Apply options:
-
If --coverage : Enable coverage reporting for all tiers
-
If --parallel : Run tests in parallel where safe
-
If --report : Generate HTML report
Stop on failure at any tier (don't waste time on later tiers)
Provide pyramid summary:
Full Test Suite: [PASSED|FAILED]
| Tier | Passed | Failed | Duration |
|---|---|---|---|
| Unit | X | Y | Zs |
| Integration | X | Y | Zs |
| E2E | X | Y | Zs |
Coverage: XX% (target: 80%)
Failures
[Grouped by tier with file:line references]
Recommended Actions
-
[Specific next steps]
Post-action guidance:
-
All pass: Ready to commit/PR
-
Unit failures: Fix immediately, use /test:quick for iteration
-
Integration failures: Check service boundaries
-
E2E failures: Check selectors/timing
-
Coverage gaps: Use /test:consult coverage
Provide the agent with:
-
All context from the section above
-
The parsed parameters
-
Explicit instruction: Run all tiers in order
The agent has expertise in:
-
Full test pyramid execution
-
Coverage analysis and reporting
-
E2E test frameworks (Playwright, Cypress)
-
CI/CD integration
Agent Teams (Optional)
For large test suites, spawn teammates for parallel test execution:
Teammate Focus Value
Unit test runner Fast unit tests Quick feedback loop, fail-fast
Integration test runner Component interactions Service boundary validation
E2E test runner Full user flows End-to-end verification
Each teammate runs its tier independently and reports results via the shared task list. This is optional — the skill runs tiers sequentially without agent teams.