renovate-review

Pass

Audited by Gen Agent Trust Hub on May 16, 2026

Risk Level: SAFE
Full Analysis
  • [PROMPT_INJECTION]: The skill identifies a potential attack surface for indirect prompt injection as it processes untrusted content from GitHub PRs and external documentation.\n
  • Ingestion points: PR title, body, and file list via gh pr view and external documentation via WebFetch (SKILL.md).\n
  • Boundary markers: Absent; there are no specific instructions to the agent to disregard instructions found within the PR content or changelogs.\n
  • Capability inventory: The skill can execute local commands (Bash, Grep, Glob) and perform network writes (gh pr comment).\n
  • Sanitization: Absent; external content is used directly to populate the safety assessment report report template.\n- [EXTERNAL_DOWNLOADS]: The skill fetches documentation from GitHub to identify breaking changes.\n
  • Evidence: Uses WebFetch to read release pages and changelogs from https://github.com/<owner>/<repo>/ (SKILL.md).\n
  • Note: These requests target a well-known service (GitHub) to perform the skill's primary function of dependency review.
Audit Metadata
Risk Level
SAFE
Analyzed
May 16, 2026, 05:39 AM
Security Audit — agent-trust-hub — renovate-review