you-web
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and process untrusted external data from the web using search and content extraction tools.
- Ingestion points: Data enters the agent's context through the
you-search(snippets) andyou-contents(full-page markdown/HTML) tools. - Boundary markers: The skill includes explicit instructions for the agent to ignore any instructions found within
<external-content>blocks. - Capability inventory: The skill uses tools to perform network requests to search the web and retrieve page contents.
- Sanitization: The instructions explicitly direct the agent to treat all web content as untrusted evidence and to use it only as data to support claims, never as instructions to follow.
Audit Metadata