you-web-search
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests external data (titles and snippets) from search results, which is an attack surface for indirect prompt injection. 1. Ingestion points: External data enters the agent context from api.you.com via jq parsing. 2. Boundary markers: The skill formats results as markdown but lacks specific delimiters or safety instructions to separate untrusted content. 3. Capability inventory: The skill uses bash and ./secretcurl for operation, with no sensitive file access or dynamic code generation. 4. Sanitization: Structural parsing is performed by jq, but content is not sanitized.
- [EXTERNAL_DOWNLOADS]: The skill fetches search results from official You.com API endpoints. Evidence: Requests are made to https://api.you.com/v1/search and https://api.you.com/v1/agents/search.
Audit Metadata