--- name: ensemble-release description: Automated release workflow orchestration with quality gates, smoke test integration, (Codex skill for /ensemble:release) user-invocable: true --- # Ensemble Command: /ensemble:release This Codex skill mirrors the Ensemble slash command `/ensemble:release`. Follow the workflow below, adapt to the current repository, and keep outputs structured. This command orchestrates the complete release workflow from quality gates through deployment and verification. It delegates to release-agent for workflow orchestration, coordinating with specialized agents (code-reviewer, test-runner, playwright-tester, deployment-orchestrator) to ensure safe, reliable releases with comprehensive smoke test coverage and automated rollback. ## Workflow ### Phase 1: Release Initialization **1. Version Validation** Validate semantic version format (X.Y.Z) - Parse --version argument - Validate semantic versioning format - Check for version conflicts (tag/branch exists) **2. Release Type Validation** Validate release type and workflow selection - Parse --type argument (standard, hotfix, rollback) - Determine base branch (main, develop, production) - Determine target branch (release/vX.Y.Z, hotfix/vX.Y.Z) - Validate branch arguments if provided **3. Release Branch Creation** Create release branch via git-workflow - Create release branch from base - Push to remote - Verify branch created successfully **Delegation:** @git-workflow Base and target branches, release version **4. Changelog Generation** Generate changelog from git history - Parse commits since last release - Categorize changes (breaking, new, enhancement, bugfix) - Generate markdown formatted changelog - Add changelog to release branch ### Phase 2: Quality Gates **1. Security Scan** Execute security vulnerability scan via code-reviewer - Scan for OWASP Top 10 vulnerabilities - Check for critical and high-severity issues - Generate security report - Block on critical/high issues **Delegation:** @code-reviewer Release branch, security scan operation **2. Definition of Done Validation** Validate all DoD categories via code-reviewer - Validate all 8 DoD categories - Check test coverage targets (unit ≥80%, integration ≥70%) - Verify documentation updated - Block on any category failure **Delegation:** @code-reviewer Release branch, DoD validation operation **3. Unit Test Execution** Execute unit test suite via test-runner - Run unit test suite with coverage - Require ≥80% coverage - Provide intelligent failure triage - Block on failures or coverage below threshold **Delegation:** @test-runner Release branch, unit tests operation **4. Integration Test Execution** Execute integration test suite via test-runner - Run integration test suite with coverage - Require ≥70% coverage - Provide intelligent failure triage - Block on failures or coverage below threshold **Delegation:** @test-runner Release branch, integration tests operation **5. Pre-Release Smoke Tests** Execute smoke test suite before deployment - Execute 5-category smoke test suite - API Health - Database connectivity - External services - Authentication - Critical paths - Use pre-release environment configuration - Block on any smoke test failure **6. E2E Test Execution** Execute critical user journeys via playwright-tester - Execute critical user journey tests - Capture trace artifacts for all journeys - Capture screenshots on failures - Block on any journey failure **Delegation:** @playwright-tester Release branch, staging environment ### Phase 3: Staging Deployment **1. Deploy to Staging** Deploy release to staging environment - Execute blue-green or rolling deployment - Verify health checks - Route traffic to new version **Delegation:** @deployment-orchestrator Release version, staging environment, deployment strategy **2. Post-Staging Smoke Tests** Verify staging deployment with smoke tests - Execute full 5-category smoke test suite - Use staging-specific configuration - Verify all smoke tests pass - Block on any smoke test failure ### Phase 4: Production Deployment **1. Deploy to Production (Canary)** Deploy to production using canary strategy - Deploy to canary infrastructure - Route 5% traffic to canary - Execute canary smoke tests (smoke-test-runner skill) - Monitor error rate for 2 minutes - If pass: Route 25% traffic, execute smoke tests - If pass: Route 100% traffic, execute final smoke tests - If fail at any stage: Trigger automated rollback **Delegation:** @deployment-orchestrator Release version, production environment, canary strategy **2. Post-Production Smoke Tests** Final verification of production deployment - Execute full 5-category smoke test suite - Use production-specific configuration - Monitor error rate for 5 minutes - Trigger rollback on smoke test failure or error rate >5% ### Phase 5: Release Completion **1. Create Pull Request** Create release PR via github-specialist - Create PR with changelog in description - Add labels (release, version) - Request reviews from tech-lead - Link to release artifacts **Delegation:** @github-specialist Release branch, main branch, changelog **2. Create GitHub Release** Publish GitHub release with notes - Create GitHub release with tag - Attach release notes - Mark as draft if --draft flag provided **Delegation:** @github-specialist Release version, changelog, release notes **3. Generate Release Report** Create comprehensive release report - Compile test execution history - Include smoke test results - Add deployment timeline - Summarize quality gate metrics **4. Append Audit Log** Record release in audit log - Append audit log entry with test tracking - Include approval history - Record deployment events - Add release artifacts links **5. Update Tickets** Update Linear/Jira tickets with release info - Update ticket status to deployed - Attach release report - Link to PR and GitHub release - Add deployment timestamp **6. Send Release Metrics** Report metrics to manager-dashboard-agent - Send cycle time metric - Send success/failure status - Send test execution metrics - Send deployment duration **Delegation:** @manager-dashboard-agent Release metrics, cycle time, success rate ### Phase 6: Rollback (On Failure) **1. Trigger Rollback** Initiate automated rollback workflow - Detect failure condition (smoke test, error rate, health check) - Alert on-call engineer - Capture rollback trigger context **2. Revert Traffic** Route traffic to previous version - Route all traffic to previous version - Verify previous version health checks - Capture traffic reversion metrics **Delegation:** @deployment-orchestrator Rollback operation, previous version **3. Post-Rollback Smoke Tests** Verify rollback success with smoke tests - Execute full 5-category smoke test suite - Verify error rate normalized - Confirm system stability - Escalate on smoke test failure **4. Health Validation** Monitor system health post-rollback - Monitor error rate (<1% target) - Verify all health checks passing - Confirm no degradation - Escalate on health check failure **5. Create Git Revert** Create revert commit in git - Create revert commit - Tag rollback version (vX.Y.Z-rollback) - Update changelog with rollback entry **Delegation:** @git-workflow Rollback operation, release version **6. Update Tickets** Document rollback in tickets - Update ticket status to rolled-back - Add rollback reason - Link to rollback artifacts - Schedule post-mortem ## Expected Output **Format:** Release workflow results **Structure:** - **Quality Gate Results**: Security scan, DoD validation, test results (unit, integration, smoke, E2E) - **Deployment Results**: Staging deployment, production deployment (canary progression), smoke test results - **Release Artifacts**: Pull request URL, GitHub release URL, release report, audit log - **Rollback Results**: (If triggered) Traffic reversion, post-rollback smoke tests, health validation ## Usage ``` /ensemble:release ```