QE Workflows
Plan tests, execute all layers, publish results.
Two workflows that cover the full QE lifecycle — from test planning against story requirements to multi-layer execution with failure classification and Jira publishing.
The Problem
QE execution is manual, fragmented, and disconnected from planning. Test plans live in docs, execution lives in CI, and results live in spreadsheets. Regressions slip through, coverage gaps go unnoticed, and nobody knows the real quality state.
No standalone QE execution
Tests run in CI pipelines but QE can’t trigger a structured execution pass on demand.
Manual regression tracking
Failures are triaged manually — no automatic classification of regressions vs. pre-existing vs. flaky.
Planning ≠ execution gap
Test plans from story QA don’t flow into execution — QE re-discovers what to test every time.
Two workflows. Plan → Execute.
Story QA plans the tests during story readiness. QE Run executes them after development, classifies every failure, and publishes results to Jira.
Story QA
Plan · 6 phases
Plans tests across 5 layers from story ACs, Confluence API specs, and PRD edge cases. Publishes the test plan to Jira and creates QE sub-tasks per layer.
/story-qa DO-3238QE Run
Execute · 12 phases
Executes tests across 6 layers, creates missing test files and updates existing ones with new assertions, classifies failures against baseline, and publishes results to Jira.
/qe-run DO-3238Full QE Pipeline
/story-qa plans tests during readiness. Development builds the feature. /qe-run verifies after dev is complete.
Workflow Comparison
| Aspect | Story QA | QE Run |
|---|---|---|
| Role | QE / SDET (planning) | QE / SDET (execution) |
| When | Before development (story readiness) | After development (verification) |
| Input | Engineered story + Confluence specs | Jira ticket + TEST_PLAN.md (optional) |
| Output | TEST_PLAN.md + TEST_DATA_PLAN.md | QE_REPORT.md + Jira bug tickets |
| Phases | ~6 phases | ~12 phases |
| Test layers | 5 (unit, integration, E2E, security, performance) | 6 (+ accessibility) |
| Signal | STORY_SPRINT_READY | QE_COMPLETE |
How it works
Generate story workflows
Run the story generator to create /story-qa — this creates the test planning workflow used during story readiness.
Generate QE workflows
Run the QE generator to create /qe-run — it surveys your test stack, detects test repo location (same repo, monorepo, or separate), and generates a validated execution workflow in 3 phases (survey, generate, validate).
Plan with /story-qa
During story readiness, QE runs /story-qa to plan tests across 5 layers. The output (TEST_PLAN.md) is ready for execution once development completes.
Development builds the feature
Dev uses /bug-fix, /feature-enhance, or /feature-build to implement the story. The development workflow produces code changes that QE will verify in the next step.
Verify with /qe-run
After development, QE runs /qe-run to execute all test layers, classify failures against baseline, generate a report, and publish results to Jira with auto-created bug tickets.