classify-code-review

Pass

Audited by Gen Agent Trust Hub on Oct 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data including Git diffs, PR metadata, issue descriptions, and project specifications.
  • Ingestion points: Data enters the agent's context through AGENTS.md, PR metadata, Git status/diffs, task text, and SDLC artifacts (issues, specs, postmortems).
  • Boundary markers: The instructions do not explicitly mandate the use of delimiters or "ignore" markers when interpolating untrusted data into prompts or tool inputs, which is standard for review tasks.
  • Capability inventory: The skill utilizes tools.classify, performs Git operations (status, diff), and reads local files.
  • Sanitization: No specific sanitization or filtering logic is prescribed for handling external content before it is passed to classification tools, though the skill notes that embedded URLs are not fetched by the tool.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 11, 2026, 08:17 PM