web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to audit user-provided UI components and codebases. This creates an attack surface where malicious instructions embedded in the audited files (e.g., in HTML or JS comments) could potentially influence the agent's audit report or behavior.
- Ingestion points: User-specified files or the entire project directory (SKILL.md).
- Boundary markers: The instructions do not define explicit delimiters or warnings for the agent to distinguish between the audited data and the audit instructions.
- Capability inventory: The skill's primary function is to generate a text-based audit report (PASS/FAIL) and does not appear to use the audited content to drive high-risk actions like network requests or shell execution.
- Sanitization: There are no instructions for sanitizing or filtering input data before processing.
Audit Metadata