tavily
Pass
Audited by Gen Agent Trust Hub on Sep 23, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill provides tools (tavily_search, tavily_crawl, tavily_extract) that ingest untrusted content from the web into the agent context, creating a surface for indirect prompt injection. \n
- Ingestion points: Web content retrieved via tools defined in SKILL.md. \n
- Boundary markers: No explicit delimiters or instructions are provided to separate external web data from agent instructions. \n
- Capability inventory: The skill uses npx for shell command execution and provides tools for web navigation and data retrieval. \n
- Sanitization: No explicit sanitization or instruction filtering is mentioned for the fetched data. \n- [EXTERNAL_DOWNLOADS]: The skill downloads and executes the @useclawlink/cli package from the NPM registry. \n
- Evidence: Multiple npx @useclawlink/cli commands are documented in SKILL.md for setup and operation. \n- [COMMAND_EXECUTION]: The skill relies on shell command execution via npx to interact with the ClawLink service. \n
- Evidence: npx commands for login, connection, and running Tavily actions.
Audit Metadata