Agent skill
gh-workflow
Trigger and monitor GitHub Actions workflows, CI/CD pipelines, and automated builds. Use when running workflows, triggering CI, deploying via GitHub Actions, downloading artifacts, or checking workflow status on github.com
Install this agent skill to your Project
npx add-skill https://github.com/majiayu000/claude-skill-registry/tree/main/skills/data/gh-workflow
SKILL.md
gh-workflow
Run GitHub workflows on github.com
Workflow
When the user asks to run a workflow:
- Verify git remotes and confirm commits are pushed (MANDATORY):
- Check remotes:
git remote -v | grep github.com - Get current commit:
git rev-parse HEAD - Verify commit exists on remote:
git fetch github && git branch -r --contains $(git rev-parse HEAD) | grep -q github/ - If not found, push first:
git push github HEAD:main(or appropriate branch name) - Quick check:
git fetch github && git branch -r --contains HEAD | grep -q github/ || echo "WARNING: HEAD not pushed to github"
- Check remotes:
- Check the workflow file in
.github/workflows/to understand its trigger type and inputs - For
workflow_dispatchworkflows: trigger directly withgh workflow run - For
pushworkflows: push to the remote first, then the workflow runs automatically - Use appropriate parameters based on the workflow's input definitions
- For testing/iteration with matrix builds: prefer x64 over arm64 (faster provisioning)
- Download workflow artifacts to verify outputs
Usage
# List workflows
GH_HOST=github.com gh workflow list
# Run a workflow
GH_HOST=github.com gh workflow run <workflow-name>
# Run with inputs
GH_HOST=github.com gh workflow run <workflow-name> -f key=value
# View recent runs
GH_HOST=github.com gh run list --workflow=<workflow-name>
# Watch a run
GH_HOST=github.com gh run watch <run-id>
# Download artifacts from a run
GH_HOST=github.com gh run download <run-id>
Testing matrix builds
For workflows with matrix builds, test with the fastest variant first:
- Prefer x64 over arm64 (x64 runners provision faster)
- Prefer ubuntu over macos (faster startup)
- Use workflow inputs to target specific matrix cells if supported
Examples
# Run workflow_dispatch workflow with inputs
GH_HOST=github.com gh workflow run luajit.yml -f release_tag=2025.11.23 -f create_release=true
# Run workflow at specific commit (preferred - explicit and reproducible)
GH_HOST=github.com gh workflow run luajit.yml --ref $(git rev-parse HEAD)
# Or with explicit SHA
GH_HOST=github.com gh workflow run luajit.yml --ref abc1234
# Run without creating release (testing)
GH_HOST=github.com gh workflow run nvim.yml -f create_release=false
# Download artifacts from latest run
GH_HOST=github.com gh run list --workflow=luajit.yml --limit 1 --json databaseId --jq '.[0].databaseId' | xargs -I {} GH_HOST=github.com gh run download {}
Recommended Agent Skills
Expand your agent's capabilities with these related and highly-rated skills.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
agent-ops-spec
Manage specification documents in .agent/specs/. Use when user provides requirements, acceptance criteria, or feature descriptions that need to be tracked and validated against implementation.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-testing
Test strategy, execution, and coverage analysis. Use when designing tests, running test suites, or analyzing test results beyond baseline checks.
agent-ops-state
Maintain .agent state files. Use at session start, after meaningful steps, and before concluding: read/update constitution/memory/focus/issues/baseline consistently.
Didn't find tool you were looking for?