tavily
Pass
Audited by Gen Agent Trust Hub on Mar 29, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the 'orth' CLI tool to execute structured commands against the Tavily API for search and research tasks.
- [PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection because it retrieves untrusted content from the web and presents it to the agent. This is inherent to the functionality of any search or crawling tool. Ingestion points: Web content is retrieved from the search, extract, map, and crawl endpoints. Boundary markers: No delimiters or instructions to ignore embedded commands are specified in the markdown. Capability inventory: The agent can execute API requests using the 'orth' CLI. Sanitization: There is no documentation of content filtering or sanitization being applied to the external data.
Audit Metadata