status-report
Pass
Audited by Gen Agent Trust Hub on Oct 26, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources including pull request descriptions, review threads, and commit messages during the evidence-gathering phase.
- Ingestion points: PR metadata, review comments, and commit history are read from the repository using git and gh (documented in SKILL.md).
- Boundary markers: The skill does not specify the use of delimiters or ignore embedded instructions markers when processing or presenting this external text.
- Capability inventory: The skill is limited to read-only operations using git and gh, and rendering output via Mermaid or Markdown. It explicitly forbids file writes or mutations (documented in the Read-only contract in SKILL.md).
- Sanitization: No explicit sanitization or filtering of external content is mentioned.
- [SAFE]: The skill implements a strict Read-only contract, explicitly forbidding any destructive or state-changing operations such as fetching, pulling, editing files, or pushing commits.
Audit Metadata