tavily-web

Warn

Audited by Gen Agent Trust Hub on Apr 14, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The installation process involves downloading logic from an external community GitHub repository (BenedictKing/tavily-web). This source is not officially verified.
  • [COMMAND_EXECUTION]: The skill instructs the user to execute a shell command (npx skills add -g BenedictKing/tavily-web) to install the external code into the agent's environment.
  • [PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection due to its core functionality.
  • Ingestion points: External data is ingested from the web through Tavily search results, crawling, and content extraction tools as described in SKILL.md.
  • Boundary markers: The provided documentation does not indicate the use of boundary markers or instructions to ignore embedded commands within fetched web content.
  • Capability inventory: The skill possesses web crawling and search capabilities, which naturally involve interacting with untrusted third-party data.
  • Sanitization: There is no evidence in the provided file of sanitization or filtering of external content before it is processed by the agent.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Apr 14, 2026, 06:21 PM