developing-github-actions
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill is explicitly designed to improve the security posture of GitHub Actions workflows. It provides detailed references on threat modeling, credential management, and supply-chain security (e.g., in
references/security.md).\n- [INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it is designed to ingest and analyze potentially untrusted GitHub Actions workflow YAML files.\n - Ingestion points: The skill reads workflow files from the
.github/workflows/directory as part of its primary inspection and verification workflow (defined inSKILL.mdandscripts/verify.sh).\n - Boundary markers: The instructions include robust warnings and guidelines in
references/security.mdandSKILL.mdspecifically aimed at safely handling untrusted data and preventing script execution from workflow expressions.\n - Capability inventory: The skill includes a bash script (
scripts/verify.sh) that executes local binaries (actionlint,zizmor) and utilizes the GitHub CLI (gh) for repository interactions.\n - Sanitization: The skill mandates the use of environment variables for interpolating untrusted data and relies on automated security scanners to validate workflow integrity.
Audit Metadata