frontend-guidelines
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill provides instructions to download updated guidelines and commit metadata from Vercel Labs' official GitHub repository (
vercel-labs/web-interface-guidelines). This is an expected maintenance operation fetching documentation from a well-known, trusted source. - [INDIRECT_PROMPT_INJECTION]: The skill's primary function is to ingest and analyze untrusted source code provided by users. This presents an attack surface for indirect prompt injection where the data being processed contains commands targeted at the auditing agent.
- Ingestion points: Project source code files resolved from caller arguments, file lists, or patterns in the
SKILL.mdprocess. - Boundary markers: Absent. There are no specified delimiters or instructions to the model to treat file content exclusively as data to be analyzed rather than instructions to be followed.
- Capability inventory: The skill possesses the ability to read arbitrary project files and execute network requests via
curl(for the guideline refresh process). - Sanitization: No sanitization, escaping, or filtering of the audited code content is documented.
Audit Metadata