github-release-management

Pass

Audited by Gen Agent Trust Hub on Feb 28, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Employs standard gh, git, and npm commands to automate the software release lifecycle, including tagging and deployment.
  • [EXTERNAL_DOWNLOADS]: Dynamically executes the claude-flow tool via npx from the npm registry, which is a core component of the skill.
  • [PROMPT_INJECTION]: Potential for indirect prompt injection exists in the automated changelog generation from external commit messages.
  • Ingestion points: Git commit messages and PR titles fetched from the GitHub API.
  • Boundary markers: Missing markers to distinguish between metadata and AI instructions.
  • Capability inventory: Includes the ability to write to repositories, publish packages, and run local build scripts.
  • Sanitization: No sanitization is applied to the retrieved metadata before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 28, 2026, 10:51 AM