review-guide

Pass

Audited by Gen Agent Trust Hub on Nov 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from external sources such as pull request diffs, PR comments, and issue tracking systems (Jira, Linear).
  • Ingestion points: Remote diff content, existing PR comments, and ticket descriptions in SKILL.md.
  • Boundary markers: There are no explicit instructions to use delimiters or to ignore instructions found within the processed data.
  • Capability inventory: The agent determines pull request verdicts and generates walkthrough guides based on the ingested data, influencing the user's development workflow.
  • Sanitization: No mechanisms for filtering or escaping instructions found in external data are described.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 17, 2026, 01:42 AM