web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Oct 1, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches guidelines and configuration from Vercel Labs' official GitHub repository.
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest external content and treat it as a set of instructions for the agent to execute during a code review.
- Ingestion points: Retrieves markdown content from a remote repository specified in the
Guidelines Sourcesection ofSKILL.md. - Boundary markers: No explicit delimiters or instructions are provided to the agent to prevent the fetched content from overriding its internal logic.
- Capability inventory: The skill reads user-provided files or patterns to perform audits.
- Sanitization: There is no evidence of sanitization or validation of the fetched guidelines before they are used to drive the agent's behavior.
Audit Metadata