doc-comments
Pass
Audited by Gen Agent Trust Hub on Oct 12, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to fetch, review, and modify source code documentation comments. Since it processes external code content, there is a potential surface for indirect prompt injection if malicious instructions were embedded in the source code being analyzed. The skill includes instructions to limit changes during specific modes, which helps mitigate accidental behavior changes.
- Instructional Constraints: The guidelines use emphatic language (e.g., "under any circumstances", "YOU MUST NOT") to enforce specific formatting and linting rules (such as avoiding certain product terms). These are standard steering instructions intended to ensure the agent's output meets documentation standards and passes automated checks.
Audit Metadata