tavily-web

Warn

Audited by Gen Agent Trust Hub on Feb 27, 2026

Risk Level: MEDIUMEXTERNAL_DOWNLOADSPROMPT_INJECTION
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill is installed using the command npx skills add -g BenedictKing/tavily-web, which downloads code from a non-trusted community GitHub repository. This introduces the risk of executing unverified scripts or configurations on the host system.
  • [PROMPT_INJECTION]: As a tool designed for web search, content extraction, and crawling, this skill is a high-risk surface for indirect prompt injection attacks where an external website could provide malicious instructions to the agent.
  • Ingestion points: Untrusted data enters the agent's context through web search results and content crawled from external URLs (SKILL.md).
  • Boundary markers: There are no specified delimiters or 'ignore' instructions mentioned to prevent the agent from obeying commands embedded in the search results.
  • Capability inventory: The skill allows the agent to perform active web research and content scraping, expanding its knowledge base with potentially adversarial data.
  • Sanitization: The documentation does not describe any sanitization, filtering, or validation processes for the data retrieved from the web before it is presented to the model.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Feb 27, 2026, 08:41 AM