web-design-guidelines
Warn
Audited by Gen Agent Trust Hub on Nov 1, 2026
Risk Level: MEDIUMMETADATA_POISONINGEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [METADATA_POISONING]: The skill metadata identifies its author as 'vercel', which contradicts the provided developer context of 'fritzgeraldz'. This discrepancy is misleading regarding the skill's true origin.\n- [EXTERNAL_DOWNLOADS]: Fresh guidelines are fetched from a remote GitHub repository managed by a well-known organization (vercel-labs).\n- [INDIRECT_PROMPT_INJECTION]: The skill processes remote markdown content to obtain rules and output instructions, which creates a potential surface for indirect prompt injection.\n
- Ingestion points: Remote content fetched from GitHub and user-provided code files.\n
- Boundary markers: No delimiters or boundary instructions are used to separate remote instructions from the agent's core guidelines.\n
- Capability inventory: The skill utilizes network fetching (WebFetch) and file reading capabilities.\n
- Sanitization: The skill does not describe any sanitization or validation of the fetched markdown instructions.
Audit Metadata