web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
- EXTERNAL_DOWNLOADS (LOW): The skill fetches guidelines from a remote URL (https://raw.githubusercontent.com/vercel-labs/web-interface-guidelines/main/command.md). Per [TRUST-SCOPE-RULE], this is downgraded to LOW/INFO because vercel-labs is a trusted organization.
- INDIRECT_PROMPT_INJECTION (LOW): The skill processes user-provided files alongside fetched instructions, creating a surface for indirect prompt injection. Evidence Chain: 1. Ingestion points: User-specified files and the remote command.md file. 2. Boundary markers: Absent. 3. Capability inventory: Local file reading and network operations via WebFetch. 4. Sanitization: Absent.
- DYNAMIC_EXECUTION (LOW): The agent's logic and output formatting are determined by the content of the fetched command.md file at runtime. While legitimate for a linter, this constitutes dynamic instruction loading.
Audit Metadata