setup-coding-standards
Pass
Audited by Gen Agent Trust Hub on Oct 20, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill performs routine repository maintenance by generating a coding standards file based on a provided template and local code analysis. It does not perform any network operations or access sensitive credentials.
- [INDIRECT_PROMPT_INJECTION]: The skill analyzes the repository's codebase to identify and document existing coding conventions, which introduces an indirect injection surface.
- Ingestion points: The skill reads repository code files in Step 3 to identify and verify conventions.
- Boundary markers: No specific delimiters or safety instructions are defined for the input source code being analyzed.
- Capability inventory: The skill utilizes shell command execution via
git rev-parseand performs file-write operations to the repository root. - Sanitization: The instructions explicitly require the agent to verify conventions against code behavior and state that "a comment is not evidence," which acts as a validation step against potentially misleading or malicious content within code comments.
Audit Metadata