agentskillexchange Prevent broken GitHub Actions workflows before CI runs with actionlint
Use actionlint when an agent needs to inspect GitHub Actions workflow files before a push or pull request lands. The skill checks syntax, expressions, action inputs, runner labels, cron patterns, and a few security footguns so the agent can stop bad workflow changes before CI burns time.
Prerequisites
actionlint binary, plus optional shellcheck and pyflakes for deeper inline script checks
Installation
Use the upstream install or setup path that matches your environment:
- go install github.com/rhysd/actionlint/cmd/actionlint@latest
Requirements and caveats from upstream:
- uses: actions/setup-node@v4
- key: ${{ matrix.platform }}-node-${{ hashFiles('**/package-lock.json') }}
- test.yaml:17:11: input "node_version" is not defined in action "actions/setup-node@v4". available inputs are "always-auth", "architecture", "cache", "cache-dependency-path", "check-latest", "node-version", "node-versi...
Basic usage or getting-started notes:
Actions usage check to check that inputs at with: and outputs in steps.{id}.outputs are correct
[shellcheck][] and [pyflakes][] integrations for scripts at run:
Example of broken workflow:
Extracted from upstream docs: https://raw.githubusercontent.com/rhysd/actionlint/HEAD/README.md