tavily-web

Pass

Audited by Gen Agent Trust Hub on Apr 11, 2026

Risk Level: SAFEPROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it processes data from the public web.
  • Ingestion points: Web search results and URL content extraction (SKILL.md).
  • Boundary markers: The documentation does not specify delimiters to isolate untrusted web content from agent instructions.
  • Capability inventory: Ingests and processes external web content.
  • Sanitization: No sanitization or validation of retrieved content is documented.
  • [EXTERNAL_DOWNLOADS]: The installation command fetches the skill from a remote repository.
  • Source: BenedictKing/tavily-web.
  • Method: npx skills add. This is the standard mechanism for skill deployment and is considered safe in this context.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 11, 2026, 06:19 PM
Security Audit — agent-trust-hub — tavily-web