
Agent Release Swarm
Orchestrate a multi-step software release with AI swarms—from changelog and GitHub PR work through coordinated deployment hooks—when you are ready to ship a version.
Overview
Agent Release Swarm is an agent skill most often used in Ship (also Operate, Grow) that orchestrates complex releases with AI swarms, GitHub MCP, and claude-flow parallel execution.
Install
npx skills add https://github.com/ruvnet/ruflo --skill agent-release-swarmWhat is this skill?
- Coordinates release-swarm mode via npx ruv-swarm hooks: pre-task, post-edit, post-task, notification
- Wires GitHub MCP tools for branches, PRs, merge, push_files, and issues
- Uses claude-flow MCP for swarm_init, agent_spawn, task_orchestrate, parallel_execute, load_balance
- TodoWrite/TodoRead plus Task and WebFetch for cross-agent release planning
- 4 documented ruv-swarm hook modes: pre-task, post-edit, post-task, notification
Adoption & trust: 641 installs on skills.sh; 58.5k GitHub stars; 1/3 security scanners passed (skills.sh audits).
What problem does it solve?
Shipping a version means too many manual steps—changelog, PRs, branches, and deployment—and your single agent thread cannot safely parallelize them.
Who is it for?
Technical solo founders who already use ruv-swarm, GitHub MCP, and claude-flow and want agent-driven release trains beyond copy-paste CI snippets.
Skip if: First-time shippers without MCP GitHub tokens, or teams that only need a one-file CHANGELOG edit without multi-agent coordination overhead.
When should I use this skill?
Orchestrate complex software releases using AI swarms that handle everything from changelog generation to multi-platform deployment.
What do I get? / Deliverables
A coordinated release swarm finishes orchestrated tasks, syncs state on edits, and runs finalize-release and broadcast hooks after deployment-oriented work.
- Coordinated release PR and branch operations via GitHub MCP
- Swarm-synchronized release task completion with notification broadcast
Recommended Skills
Journey fit
Spans multiple journey phases - primary shelf plus alternate fits below.
Primary shelf is Ship launch because the skill finalizes releases, merges, and broadcast hooks rather than day-to-day feature coding. Launch subphase matches changelog generation, multi-platform deployment coordination, and post-release swarm notification in the documented hooks.
Where it fits
Spawn a release swarm to open a version PR, merge after checks, and run finalize-release hooks.
Re-run release-swarm mode for a patch train with post-edit validation across swarm agents.
Broadcast release completion notifications so downstream content or support agents pick up the new version.
How it compares
Use this swarm coordination skill instead of a lone writing-plans pass when release work must fan out across GitHub and parallel agents.
Common Questions / FAQ
Who is agent-release-swarm for?
Indie builders and small dev teams using ruflo/ruv-swarm with GitHub and claude-flow MCP who want AI agents to own release choreography end to end.
When should I use agent-release-swarm?
At Ship launch when tagging a release needs changelog, PR merge, and deployment coordination; at Operate iterate when hotfix trains repeat; and at Grow when release comms must broadcast across agents.
Is agent-release-swarm safe to install?
It requests Bash, git, GitHub write APIs, and swarm orchestration—high blast radius; review the Security Audits panel on this page and dry-run on a fork before production merge rights.
SKILL.md
READMESKILL.md - Agent Release Swarm
--- name: release-swarm description: Orchestrate complex software releases using AI swarms that handle everything from changelog generation to multi-platform deployment type: coordination color: "#4ECDC4" tools: - Bash - Read - Write - Edit - TodoWrite - TodoRead - Task - WebFetch - mcp__github__create_pull_request - mcp__github__merge_pull_request - mcp__github__create_branch - mcp__github__push_files - mcp__github__create_issue - mcp__claude-flow__swarm_init - mcp__claude-flow__agent_spawn - mcp__claude-flow__task_orchestrate - mcp__claude-flow__parallel_execute - mcp__claude-flow__load_balance hooks: pre_task: | echo "🐝 Initializing release swarm coordination..." npx ruv-swarm hook pre-task --mode release-swarm --init-swarm post_edit: | echo "🔄 Synchronizing release swarm state and validating changes..." npx ruv-swarm hook post-edit --mode release-swarm --sync-swarm post_task: | echo "🎯 Release swarm task completed. Coordinating final deployment..." npx ruv-swarm hook post-task --mode release-swarm --finalize-release notification: | echo "📡 Broadcasting release completion across all swarm agents..." npx ruv-swarm hook notification --mode release-swarm --broadcast --- # Release Swarm - Intelligent Release Automation ## Overview Orchestrate complex software releases using AI swarms that handle everything from changelog generation to multi-platform deployment. ## Core Features ### 1. Release Planning ```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 "$MERGED_PRS" \ --analyze-commits \ --suggest-version \ --identify-breaking \ --generate-timeline ``` ### 2. Automated Versioning ```bash # Smart version bumping npx ruv-swarm github release-version \ --strategy "semantic" \ --analyze-changes \ --check-breaking \ --update-files ``` ### 3. Release Orchestration ```bash # Full release automation with gh CLI # Generate changelog from PRs and commits CHANGELOG=$(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 "$CHANGELOG" \ --label "announcement,release" ``` ## Release Configuration ### Release Config File ```yaml # .github$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: git