tavily-extract

Pass

Audited by Gen Agent Trust Hub on Jul 21, 2026

Risk Level: SAFEREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructions include a quick install command curl -fsSL https://cli.tavily.com/install.sh | bash which pipes a remote script from the vendor's infrastructure directly into the shell for execution.
  • [EXTERNAL_DOWNLOADS]: The skill references and downloads an installation script from the vendor-owned domain cli.tavily.com.
  • [COMMAND_EXECUTION]: The skill uses the tvly command-line tool via the Bash tool to perform content extraction and account authentication.
  • [PROMPT_INJECTION]: The skill exhibits a surface for indirect prompt injection by retrieving and processing content from arbitrary external URLs.
  • Ingestion points: Web content retrieved from user-specified URLs.
  • Boundary markers: None present in the instructions.
  • Capability inventory: Shell access restricted to the tvly command.
  • Sanitization: Content is converted to LLM-optimized markdown or plain text by the Tavily service.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 21, 2026, 12:09 AM
Security Audit — agent-trust-hub — tavily-extract