Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Agent skill for release-swarm - invoke with $agent-release-swarm
.claude/skills/agent-release-swarm/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-03 | ✗→✓ | ▲ Improved | — | — |
| case-16 | ✗→✓ | ▲ Improved | — | — |
| case-07 | ✗→✓ | ▲ Improved | — | — |
| case-17 | ✗→✓ | ▲ Improved | — | — |
| case-08 | ✗→✓ | ▲ Improved | — | — |
Orchestrate complex software releases using AI swarms that handle everything from changelog generation to multi-platform deployment.
bash# Plan next release using gh CLI # Get commit history since last release LAST_TAG=$(gh release list --limit 1 --json tagName -q '.[0].tagName') COMMITS=$(gh api repos/:owner/:repo$compare/${LAST_TAG}...HEAD --jq '.commits') # Get merged PRs MERGED_PRS=$(gh pr list --state merged --base main --json number,title,labels,mergedAt \ --jq ".[] | select(.mergedAt > \"$(gh release view $LAST_TAG --json publishedAt -q .publishedAt)\")") # Plan release with commit analysis npx ruv-swarm github release-plan \ --commits "$COMMITS" \ --merged-prs "$MERGED_PRS" \ --analyze-commits \ --suggest-version \ --identify-breaking \ --generate-timeline
bash# Smart version bumping npx ruv-swarm github release-version \ --strategy "semantic" \ --analyze-changes \ --check-breaking \ --update-files
bash# Full release automation with gh CLI # Generate changelog from PRs and commits CHANGELOG=$(gh api repos/:owner/:repo$compare/${LAST_TAG}...HEAD \ --jq '.commits[].commit.message' | \ npx ruv-swarm github generate-changelog) # Create release draft gh release create v2.0.0 \ --draft \ --title "Release v2.0.0" \ --notes "$CHANGELOG" \ --target main # Run release orchestration npx ruv-swarm github release-create \ --version "2.0.0" \ --changelog "$CHANGELOG" \ --build-artifacts \ --deploy-targets "npm,docker,github" # Publish release after validation gh release edit v2.0.0 --draft=false # Create announcement issue gh issue create \ --title "🎉 Released v2.0.0" \ --body "$CHANGELOG" \ --label "announcement,release"
yaml# .github$release-swarm.yml version: 1 release: versioning: strategy: semantic breaking-keywords: ["BREAKING", "!"] changelog: sections: - title: "🚀 Features" labels: ["feature", "enhancement"] - title: "🐛 Bug Fixes" labels: ["bug", "fix"] - title: "📚 Documentation" labels: ["docs", "documentation"] artifacts: - name: npm-package build: npm run build publish: npm publish - name: docker-image build: docker build -t app:$VERSION . publish: docker push app:$VERSION - name: binaries build: .$scripts$build-binaries.sh upload: github-release deployment: environments: - name: staging auto-deploy: true validation: npm run test:e2e - name: production approval-required: true rollback-enabled: true notifications: - slack: releases-channel - email: stakeholders@company.com - discord: webhook-url
bash# Generate intelligent changelog with gh CLI # Get all merged PRs between versions PRS=$(gh pr list --state merged --base main --json number,title,labels,author,mergedAt \ --jq ".[] | select(.mergedAt > \"$(gh release view v1.0.0 --json publishedAt -q .publishedAt)\")") # Get contributors CONTRIBUTORS=$(echo "$PRS" | jq -r '[.author.login] | unique | join(", ")') # Get commit messages COMMITS=$(gh api repos/:owner/:repo$compare$v1.0.0...HEAD \ --jq '.commits[].commit.message') # Generate categorized changelog CHANGELOG=$(npx ruv-swarm github changelog \ --prs "$PRS" \ --commits "$COMMITS" \ --contributors "$CONTRIBUTORS" \ --from v1.0.0 \ --to HEAD \ --categorize \ --add-migration-guide) # Save changelog echo "$CHANGELOG" > CHANGELOG.md # Create PR with changelog update gh pr create \ --title "docs: Update changelog for v2.0.0" \ --body "Automated changelog update" \ --base main
Capabilities:
bash# Determine next version npx ruv-swarm github version-suggest \ --current v1.2.3 \ --analyze-commits \ --check-compatibility \ --suggest-pre-release
Logic:
bash# Coordinate multi-platform builds npx ruv-swarm github release-build \ --platforms "linux,macos,windows" \ --architectures "x64,arm64" \ --parallel \ --optimize-size
Features:
bash# Pre-release testing npx ruv-swarm github release-test \ --suites "unit,integration,e2e,performance" \ --environments "node:16,node:18,node:20" \ --fail-fast false \ --generate-report
bash# Multi-target deployment npx ruv-swarm github release-deploy \ --targets "npm,docker,github,s3" \ --staged-rollout \ --monitor-metrics \ --auto-rollback
yaml# Staged rollout configuration deployment: strategy: progressive stages: - name: canary percentage: 5 duration: 1h metrics: - error-rate < 0.1% - latency-p99 < 200ms - name: partial percentage: 25 duration: 4h validation: automated-tests - name: full percentage: 100 approval: required
bash# Coordinate releases across repos npx ruv-swarm github multi-release \ --repos "frontend:v2.0.0,backend:v2.1.0,cli:v1.5.0" \ --ensure-compatibility \ --atomic-release \ --synchronized
bash# Emergency hotfix process npx ruv-swarm github hotfix \ --issue 789 \ --target-version v1.2.4 \ --cherry-pick-commits \ --fast-track-deploy
yaml# .github$workflows$release.yml name: Release Workflow on: push: tags: ['v*'] jobs: release-swarm: runs-on: ubuntu-latest steps: - uses: actions$checkout@v3 with: fetch-depth: 0 - name: Setup GitHub CLI run: echo "${{ secrets.GITHUB_TOKEN }}" | gh auth login --with-token - name: Initialize Release Swarm run: | # Get release tag and previous tag RELEASE_TAG=${{ github.ref_name }} PREV_TAG=$(gh release list --limit 2 --json tagName -q '.[1].tagName') # Get PRs and commits for changelog PRS=$(gh pr list --state merged --base main --json number,title,labels,author \ --search "merged:>=$(gh release view $PREV_TAG --json publishedAt -q .publishedAt)") npx ruv-swarm github release-init \ --tag $RELEASE_TAG \ --previous-tag $PREV_TAG \ --prs "$PRS" \ --spawn-agents "changelog,version,build,test,deploy" - name: Generate Release Assets run: | # Generate changelog from PR data CHANGELOG=$(npx ruv-swarm github release-changelog \ --format markdown) # Update release notes gh release edit ${{ github.ref_name }} \ --notes "$CHANGELOG" # Generate and upload assets npx ruv-swarm github release-assets \ --changelog \ --binaries \ --documentation - name: Upload Release Assets run: | # Upload generated assets to GitHub release for file in dist/*; do gh release upload ${{ github.ref_name }} "$file" done - name: Publish Release run: | # Publish to package registries npx ruv-swarm github release-publish \ --platforms all # Create announcement issue gh issue create \ --title "🚀 Released ${{ github.ref_name }}" \ --body "See [release notes](https:/$github.com/${{ github.repository }}$releases$tag/${{ github.ref_name }})" \ --label "announcement"
bash# Automated deployment pipeline npx ruv-swarm github cd-pipeline \ --trigger "merge-to-main" \ --auto-version \ --deploy-on-success \ --rollback-on-failure
bash# Comprehensive validation npx ruv-swarm github release-validate \ --checks " version-conflicts, dependency-compatibility, api-breaking-changes, security-vulnerabilities, performance-regression, documentation-completeness " \ --block-on-failure
bash# Test backward compatibility npx ruv-swarm github compat-test \ --previous-versions "v1.0,v1.1,v1.2" \ --api-contracts \ --data-migrations \ --generate-report
bash# Security validation npx ruv-swarm github release-security \ --scan-dependencies \ --check-secrets \ --audit-permissions \ --sign-artifacts
bash# Monitor release health npx ruv-swarm github release-monitor \ --version v2.0.0 \ --metrics "error-rate,latency,throughput" \ --alert-thresholds \ --duration 24h
bash# Configure auto-rollback npx ruv-swarm github rollback-config \ --triggers '{ "error-rate": ">5%", "latency-p99": ">1000ms", "availability": "<99.9%" }' \ --grace-period 5m \ --notify-on-rollback
bash# Analyze release performance npx ruv-swarm github release-analytics \ --version v2.0.0 \ --compare-with v1.9.0 \ --metrics "adoption,performance,stability" \ --generate-insights
bash# Update documentation npx ruv-swarm github release-docs \ --api-changes \ --migration-guide \ --example-updates \ --publish-to "docs-site,wiki"
markdown<!-- Auto-generated release notes template --> # Release v2.0.0 ## 🎉 Highlights - Major feature X with 50% performance improvement - New API endpoints for feature Y - Enhanced security with feature Z ## 🚀 Features ### Feature Name (#PR) Detailed description of the feature... ## 🐛 Bug Fixes ### Fixed issue with... (#PR) Description of the fix... ## 💥 Breaking Changes ### API endpoint renamed - Before: `$api$old-endpoint` - After: `$api$new-endpoint` - Migration: Update all client calls... ## 📈 Performance Improvements - Reduced memory usage by 30% - API response time improved by 200ms ## 🔒 Security Updates - Updated dependencies to patch CVE-XXXX - Enhanced authentication mechanism ## 📚 Documentation - Added examples for new features - Updated API reference - New troubleshooting guide ## 🙏 Contributors Thanks to all contributors who made this release possible!
bash# NPM package release npx ruv-swarm github npm-release \ --version patch \ --test-all \ --publish-beta \ --tag-latest-on-success
bash# Docker multi-arch release npx ruv-swarm github docker-release \ --platforms "linux$amd64,linux$arm64" \ --tags "latest,v2.0.0,stable" \ --scan-vulnerabilities \ --push-to "dockerhub,gcr,ecr"
bash# Mobile app store release npx ruv-swarm github mobile-release \ --platforms "ios,android" \ --build-release \ --submit-review \ --staged-rollout
bash# Emergency hotfix npx ruv-swarm github emergency-release \ --severity critical \ --bypass-checks security-only \ --fast-track \ --notify-all
bash# Immediate rollback npx ruv-swarm github rollback \ --to-version v1.9.9 \ --reason "Critical bug in v2.0.0" \ --preserve-data \ --notify-users
See also: workflow-automation.md, multi-repo-swarm.md
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-03 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-16 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-07 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-20 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-17 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-08 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-09 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-14 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-01 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-19 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-11 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-13 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-22 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-21 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
case-15 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-10 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-18 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-02 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-04 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-06 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-05 | fail→pass | — | — | — | — | — | — | — | — | — | — | — | — |
case-12 | fail→fail | — | — | — | — | — | — | — | — | — | — | — | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted, and 21 counted toward the lift figure. The other 1 produced results that are not comparable between the two arms, so they are excluded from the headline rather than averaged into it. The headline lift of +77 percentage points is the difference between those two pass rates over the 21 comparable cases.
The per-case answers from this run were removed by the retention sweep, so the case table below shows the verdicts without the text either arm produced. The counts above were recorded at the time and are unaffected. Answers are now kept for 180 days.
Other measured skills in the registry, with their headline benchmark lift.