web-design-guidelines
Pass
Audited by Gen Agent Trust Hub on Mar 2, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: Fetches design guidelines from Vercel Labs' official GitHub repository. This targets a trusted organization and is consistent with the skill's intended purpose.
- [PROMPT_INJECTION]: The skill presents a surface for indirect prompt injection as it processes untrusted user-provided files.
- Ingestion points: Reads local files based on user-provided file paths or patterns in SKILL.md.
- Boundary markers: No explicit delimiters or instructions to ignore embedded prompts are provided to the agent for the processed content.
- Capability inventory: The skill has permissions to read local files and perform external network requests via WebFetch.
- Sanitization: There is no evidence of sanitization or filtering of the file content before it is processed by the model.
Audit Metadata