nestjs-git-commit-pr-message

Pass

Audited by Gen Agent Trust Hub on Aug 3, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill implements a robust security model for Git operations, requiring explicit authorization for all state-changing remote actions and prohibiting unguarded history overwrites.
  • [SAFE]: It mitigates data exposure by mandating a scan of the staged diff for credentials, API keys, and sensitive files before any commit is created.
  • [SAFE]: The skill addresses indirect prompt injection by instructing the agent to treat issue descriptions, pull request bodies, and commit messages fetched from remotes as untrusted data rather than instructions.
  • [SAFE]: Execution of project scripts is limited to verified repository checks like linting and testing, with a clear instruction to avoid inventing new scripts.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 3, 2026, 08:31 AM
Security Audit — agent-trust-hub — nestjs-git-commit-pr-message