mittuled regression-test-runner
Agent: Social Media Manager
L2 QA and test engineer responsible for unit, integration, regression, performance, and security testing. Validates instrumentation and staging before production deployment.
Department ethos: ideal-engineering.md
Skill Description
The QA / Test Engineer runs regression tests to ensure new changes have not broken existing functionality.
When to Use
- A feature branch is ready for merge and must pass the regression gate.
- A hotfix has been applied and existing functionality needs re-verification.
- A dependency upgrade or refactor changes shared code paths.
- The team is preparing a release candidate and needs full regression sign-off.
Workflow
- Determine the scope of regression testing based on the change set and affected modules.
- Select the appropriate regression test suite (full or targeted) based on risk and time constraints.
- Verify the test environment matches the expected configuration and data state.
- Execute the regression test suite and capture pass/fail results, execution times, and logs.
- Triage failures: categorize each as a genuine regression, a flaky test, or an environment issue.
- For genuine regressions, file defect tickets with reproduction steps, expected vs. actual behavior, and severity.
- Re-run flaky tests in isolation to confirm or dismiss them.
- Report the regression test summary with pass rate, blocking defects, and release recommendation.
- Deliverable: Regression test report with pass/fail summary, defect tickets, and release recommendation.
Anti-Patterns
- Running only a subset of regressions to save time without a risk assessment. Why: Skipping tests without understanding risk leaves blind spots that surface as production incidents.
- Treating all failures as flaky without investigation. Why: Dismissing failures as flaky normalizes ignoring real regressions and degrades suite trust.
- Not updating the regression suite after new features ship. Why: A stale regression suite provides false confidence because it does not cover recent functionality.
- Running regressions against stale test data. Why: Tests that pass on outdated data may fail on production-like data, hiding real defects.
Output
Success: A regression test report confirming no genuine regressions, with a clear release recommendation.
Failure: A blocking defect list with severity, reproduction steps, and affected features, plus a recommendation to hold the release until resolved.
Related Skills
None defined yet.
performance-tester— sibling skill under the same agent — combine with performance-tester for end-to-end coverageinstrumentation-verifier-qa— sibling skill under the same agent — combine with instrumentation-verifier-qa for end-to-end coverageunit-test-runner— sibling skill under the same agent — combine with unit-test-runner for end-to-end coverage