actions-ci-tuning
Pass
Audited by Gen Agent Trust Hub on Sep 11, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill guides the agent to inventory and audit local GitHub Actions workflow files (.github/workflows/.yml), creating a surface where malicious content in these files could influence agent behavior. 1. Ingestion points: .github/workflows/.yml files (SKILL.md). 2. Boundary markers: No explicit delimiters or instructions to ignore embedded data are present. 3. Capability inventory: Uses shell commands via the GitHub CLI (gh) for auditing (SKILL.md). 4. Sanitization: No content filtering is specified.
- [EXTERNAL_DOWNLOADS]: The skill references resources from well-known providers. It suggests using the official Playwright Docker image from Microsoft (mcr.microsoft.com) and provides verified commit SHAs for official GitHub Actions from the actions and aws-actions organizations, which promotes supply chain security.
Audit Metadata