release

Pass

Audited by Gen Agent Trust Hub on Jul 1, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill uses the GitHub CLI (gh) to run workflows, list runs, and create pull requests or releases. It also uses npm for distribution tag management and git for tag discovery.
  • [COMMAND_EXECUTION]: Uses node -p to evaluate JavaScript for extracting the version field from local package.json files.
  • [PROMPT_INJECTION]: The skill processes content from CHANGELOG.md files, which is a potential surface for indirect prompt injection.
  • Ingestion points: Package-specific CHANGELOG.md files.
  • Boundary markers: None; uses awk to extract text between version headers.
  • Capability inventory: Includes shell command execution (gh, npm, node) and GitHub API interactions.
  • Sanitization: None; the extracted changelog text is passed directly to the gh release create command.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 1, 2026, 09:41 AM
Security Audit — agent-trust-hub — release