web-search
Pass
Audited by Gen Agent Trust Hub on Oct 9, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to fetch and process untrusted content from the web, which creates a surface for indirect prompt injection.
- Ingestion points: The
fetch_url_contenttool inSKILL.mdallows the agent to read arbitrary content from any URL. - Boundary markers: The skill instructions do not specify boundary markers or include explicit instructions for the agent to ignore natural language commands found within the fetched content.
- Capability inventory: The skill provides data ingestion. While this skill itself lacks high-privilege tools, the processed content could influence an agent that possesses other capabilities like file manipulation or command execution.
- Sanitization: Although the tool description states it strips scripts and boilerplate HTML, this does not protect against natural language instructions embedded in the readable text of a page.
Audit Metadata