The Agent Skills Directory

[SAFE]: The skill consists entirely of educational content and code snippets that demonstrate secure vs. insecure patterns in GitHub Actions. No executable code or autonomous actions are present.
[SAFE]: It correctly identifies and explains critical security risks including unsafe use of pull_request_target, script injection vulnerabilities via shell interpolation, and the risks of unpinned third-party actions.
[SAFE]: Recommendations align with industry standard best practices, such as applying the principle of least privilege through explicit workflow permissions and using OIDC for cloud authentication.

actions-authoring