web-search
Pass
Audited by Gen Agent Trust Hub on Apr 9, 2026
Risk Level: SAFE
Full Analysis
- [SAFE]: The skill facilitates web search and content extraction using the
infshCLI tool and associated applications (Tavily and Exa), which are the intended services of the vendor. - [COMMAND_EXECUTION]: The skill demonstrates the use of the
infshcommand to interact with AI models and search engines. This is restricted by theallowed-toolsconfiguration in the metadata. - [PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted web content through extraction tools, creating a surface for indirect prompt injection if the output is passed to an LLM without proper sanitization.
- Ingestion points:
tavily/extract,exa/extract(defined in SKILL.md) - Boundary markers: None present in the usage examples
- Capability inventory:
infsh app run(command execution via Bash) - Sanitization: No explicit sanitization or filtering of web content is mentioned.
Audit Metadata