GitHub Workflow Automation
Comprehensive GitHub Actions automation with claude-flow swarm coordination —
CI/CD pipeline generation, workflow orchestration, and repository automation
built as self-organising, adaptive GitHub workflows.
When to use
Reach for this skill when the task is about GitHub Actions workflow files or
CI/CD pipelines: generating a new .github/workflows/*.yml, optimizing an
existing pipeline (parallelization, caching, cost), analysing failed runs,
building swarm-driven CI/security/deploy stages, or wiring claude-flow swarm
coordination into a repo's automation.
When not to use
- PR-level code review with specialised agents → github-code-review
- Release versioning, changelogs, deployment orchestration → github-release-management
- Cross-repository package synchronisation → github-multi-repo
- Issue tracking and sprint planning → github-project-management
- Non-GitHub workflow automation or cloud swarm orchestration → flow-nexus-swarm
Quick start
# Generate an optimal pipeline from the codebase
npx ruv-swarm actions generate-workflow \
--analyze-codebase \
--detect-languages \
--create-optimal-pipeline
# Optimize an existing workflow
npx ruv-swarm actions optimize \
--workflow ".github/workflows/ci.yml" \
--suggest-parallelization
# Analyze a failed run and get fixes
gh run view <run-id> --json jobs,conclusion | \
npx ruv-swarm actions analyze-failure --suggest-fixes
Claude Code only: the swarm generation/optimisation above (ruv-swarm actions) needs
the claude-flow MCP server. On Codex / GPT-6 Astra (or any session without it), inspect
and drive runs with plain gh instead: gh run list --workflow=ci.yml, gh run view <run-id> --log-failed, gh run rerun <run-id> --failed, gh workflow run ci.yml --ref main — see
../github-code-review/references/plain-gh-fallback.md
for the full set. Writing/editing the workflow YAML itself is a plain text edit either way.
Prerequisites
gh installed and authenticated · git configured · Node.js v16+ ·
claude-flow@alpha available · repo has .github/workflows and Actions
enabled. Full checklist and a setup script: references/best-practices.md.
Reference detail (progressive disclosure)
The heavy catalogs live in references/ — load the one you need:
- references/github-modes.md — the eight
swarm-powered GitHub modes (gh-coordinator, pr-manager, issue-tracker,
release-manager, repo-architect, code-reviewer, ci-orchestrator,
security-guardian) with usage examples.
- references/workflow-templates.md —
eight production-ready Actions templates (intelligent CI, polyglot detection,
adaptive security scan, self-healing pipeline, progressive deploy, perf-guard,
PR validation, intelligent release) plus monitoring/analytics commands.
- references/advanced-features.md —
dynamic test strategies, predictive analysis, custom swarm-action
development, and MCP-based claude-flow swarm coordination / hooks / batch ops.
- references/best-practices.md — workflow
organisation, security and performance best practices, debugging &
troubleshooting, complete real-world examples, the full command reference,
and the setup checklist.
Related skills
github-code-review — PR review and quality analysisgithub-release-management — release automation and deploymentgithub-project-management — issue tracking and sprint planninggithub-multi-repo — cross-repository coordination
1---2name: github-workflow-automation3description: Automate GitHub Actions workflows and CI/CD pipelines with claude-flow swarm coordination. Use when creating, optimizing, or debugging GitHub Actions workflow YAML, CI/CD pipelines, or GitHub repository automation — not for PR code review, releases, issues, or multi-repo sync (see the github-* sibling skills).4---56# GitHub Workflow Automation78Comprehensive GitHub Actions automation with claude-flow swarm coordination —9CI/CD pipeline generation, workflow orchestration, and repository automation10built as self-organising, adaptive GitHub workflows.1112## When to use1314Reach for this skill when the task is about GitHub Actions *workflow files* or15CI/CD *pipelines*: generating a new `.github/workflows/*.yml`, optimizing an16existing pipeline (parallelization, caching, cost), analysing failed runs,17building swarm-driven CI/security/deploy stages, or wiring claude-flow swarm18coordination into a repo's automation.1920## When not to use2122- PR-level code review with specialised agents → **github-code-review**23- Release versioning, changelogs, deployment orchestration → **github-release-management**24- Cross-repository package synchronisation → **github-multi-repo**25- Issue tracking and sprint planning → **github-project-management**26- Non-GitHub workflow automation or cloud swarm orchestration → **flow-nexus-swarm**2728## Quick start2930```bash31# Generate an optimal pipeline from the codebase32npx ruv-swarm actions generate-workflow \33 --analyze-codebase \34 --detect-languages \35 --create-optimal-pipeline3637# Optimize an existing workflow38npx ruv-swarm actions optimize \39 --workflow ".github/workflows/ci.yml" \40 --suggest-parallelization4142# Analyze a failed run and get fixes43gh run view <run-id> --json jobs,conclusion | \44 npx ruv-swarm actions analyze-failure --suggest-fixes45```4647**Claude Code only:** the swarm generation/optimisation above (`ruv-swarm actions`) needs48the claude-flow MCP server. On Codex / GPT-6 Astra (or any session without it), inspect49and drive runs with plain `gh` instead: `gh run list --workflow=ci.yml`, `gh run view50<run-id> --log-failed`, `gh run rerun <run-id> --failed`, `gh workflow run ci.yml --ref51main` — see52[../github-code-review/references/plain-gh-fallback.md](../github-code-review/references/plain-gh-fallback.md)53for the full set. Writing/editing the workflow YAML itself is a plain text edit either way.5455## Prerequisites5657`gh` installed and authenticated · `git` configured · Node.js v16+ ·58`claude-flow@alpha` available · repo has `.github/workflows` and Actions59enabled. Full checklist and a setup script: `references/best-practices.md`.6061## Reference detail (progressive disclosure)6263The heavy catalogs live in `references/` — load the one you need:6465- **[references/github-modes.md](references/github-modes.md)** — the eight66 swarm-powered GitHub modes (gh-coordinator, pr-manager, issue-tracker,67 release-manager, repo-architect, code-reviewer, ci-orchestrator,68 security-guardian) with usage examples.69- **[references/workflow-templates.md](references/workflow-templates.md)** —70 eight production-ready Actions templates (intelligent CI, polyglot detection,71 adaptive security scan, self-healing pipeline, progressive deploy, perf-guard,72 PR validation, intelligent release) plus monitoring/analytics commands.73- **[references/advanced-features.md](references/advanced-features.md)** —74 dynamic test strategies, predictive analysis, custom swarm-action75 development, and MCP-based claude-flow swarm coordination / hooks / batch ops.76- **[references/best-practices.md](references/best-practices.md)** — workflow77 organisation, security and performance best practices, debugging &78 troubleshooting, complete real-world examples, the full command reference,79 and the setup checklist.8081## Related skills8283- `github-code-review` — PR review and quality analysis84- `github-release-management` — release automation and deployment85- `github-project-management` — issue tracking and sprint planning86- `github-multi-repo` — cross-repository coordination
dreamlab-ai