managing-github-ci

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFE
Full Analysis
  • [Prompt Injection] (SAFE): No malicious prompt injection patterns or attempts to override agent safety protocols were detected.
  • [Data Exposure & Exfiltration] (SAFE): No hardcoded credentials or sensitive file exfiltration patterns were identified. The skill correctly demonstrates using GitHub Secrets (NPM_TOKEN, GITHUB_TOKEN) for authentication.
  • [External Downloads] (SAFE): The skill references standard, well-known GitHub Actions (e.g., actions/checkout, pnpm/action-setup) and established package managers like pnpm and npm.
  • [Remote Code Execution] (SAFE): No instances of executing remote scripts via curl | bash or unsafe dynamic code evaluation were found.
  • [Command Execution] (SAFE): Tool usage is limited to the GitHub CLI (gh) and standard project management commands (pnpm), which is appropriate for the skill's stated purpose.
  • [Indirect Prompt Injection] (SAFE): While the skill interacts with external data (PR content and logs), the risk is minimal and inherent to the functionality of CI/CD management tools.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 07:58 PM