preview-environments-builder

Pass

Audited by Gen Agent Trust Hub on Sep 4, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill consists of standard CI/CD configuration templates intended for automation. The instructions and YAML snippets align with common DevOps practices for creating preview deployments.
  • [CREDENTIALS_UNSAFE]: The skill demonstrates safe credential handling by using GitHub Secrets placeholders (e.g., ${{ secrets.VERCEL_TOKEN }}, ${{ secrets.PREVIEW_DB_URL }}) instead of hardcoding sensitive keys or passwords.
  • [EXTERNAL_DOWNLOADS]: Referenced external resources include official GitHub Actions (actions/checkout, actions/setup-node, actions/github-script) and a widely recognized community action (amondnet/vercel-action). These are expected dependencies for the stated purpose of the skill.
  • [INDIRECT_PROMPT_INJECTION]: The automation includes a step to post a PR comment containing deployment metadata. While this processes external data (PR number, deployment URL), the ingestion and usage are standard for CI/CD notifications and do not introduce significant attack surface.
  • [COMMAND_EXECUTION]: Uses standard command-line tools such as npm, docker, kubectl, and psql within the context of a build runner to perform legitimate deployment and database seeding tasks.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 4, 2026, 09:30 AM
Security Audit — agent-trust-hub — preview-environments-builder