Skills
Navigate
skills/hwj123hwj/custom-skills/tavily/Gen Agent Trust Hub

tavily

Fail

Audited by Gen Agent Trust Hub on Jul 31, 2026

Risk Level: HIGHREMOTE_CODE_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [REMOTE_CODE_EXECUTION]: The skill instructions advise installing the Tavily CLI by piping a remote shell script directly to bash (curl -fsSL https://cli.tavily.com/install.sh | bash). This allows the execution of unverified remote code on the system.
  • [EXTERNAL_DOWNLOADS]: The installation process involves fetching a script from the external domain cli.tavily.com.
  • [PROMPT_INJECTION]: The skill ingests and processes untrusted data from the web, making it susceptible to indirect prompt injection.
  • Ingestion points: Data retrieved via tvly search and tvly extract in SKILL.md is added to the agent's context.
  • Boundary markers: There are no explicit instructions or delimiters used to separate the external content from the agent's primary instructions.
  • Capability inventory: The agent uses the Bash tool to execute tvly search and research commands.
  • Sanitization: No sanitization or filtering of the retrieved web content is specified in the instructions.
Recommendations
  • HIGH: Downloads and executes remote code from: https://cli.tavily.com/install.sh - DO NOT USE without thorough review
  • AI detected serious security threats
Audit Metadata
Risk Level
HIGH
Analyzed
Jul 31, 2026, 03:23 PM