web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: LOWEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- EXTERNAL_DOWNLOADS (LOW): The skill fetches guideline content from 'https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md'. While this is a remote fetch, the source organization 'vercel-labs' is a listed Trusted External Source, which downgrades the severity per [TRUST-SCOPE-RULE].
- PROMPT_INJECTION (LOW): Category 8 (Indirect Prompt Injection) vulnerability identified. * Ingestion points: The skill fetches external guidelines and reads user-specified local UI files (SKILL.md, lines 11, 22). * Boundary markers: Absent; there are no instructions to the agent to treat the content of the UI files as untrusted data or to ignore embedded instructions. * Capability inventory: The skill is limited to reading files and outputting textual findings; no shell execution, file-write, or non-whitelisted network capabilities are present. * Sanitization: Absent; the agent is instructed to directly apply all rules found in the fetched content.
Audit Metadata