dotnet-github-releases

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill incorporates external data from CHANGELOG.md and git-cliff output into release notes, which are then processed by the gh tool. This creates a surface where malicious content in a changelog could attempt to influence agent behavior.
  • Ingestion points: Release notes extracted from CHANGELOG.md using sed and output from the git-cliff tool.
  • Boundary markers: Absent; the skill does not use specific delimiters or instructions to ignore embedded commands within the release notes body.
  • Capability inventory: The skill uses the gh CLI and GitHub REST API (curl) for network operations and release creation.
  • Sanitization: Absent; no validation or escaping of the ingested text is performed before interpolation into the release body.
  • [EXTERNAL_DOWNLOADS]: The skill utilizes the softprops/action-gh-release@v2 GitHub Action and actions/checkout@v4 for automating the release process in CI/CD environments. These are well-known community and official actions used for their intended purposes.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 07:31 AM
Security Audit — agent-trust-hub — dotnet-github-releases