tavily-extract

CRITICAL E005: Suspicious download URL detected in skill instructions.

• Suspicious download URL detected (high risk: 1.00). Includes a direct curl|bash installation script hosted on an external domain (cli.tavily.com), which is a high-risk pattern for distributing malicious executables or scripts when the source is not a well-known, trusted vendor.

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). Yes: this workflow’s required runtime step is to fetch and extract readable markdown/text from user-supplied URLs via the Tavily CLI, so the agent’s LLM context can ingest outsider-authored web page content (public web content fetched at runtime).

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The SKILL's quick-install uses a curl | bash command that fetches and executes remote code from https://cli.tavily.com/install.sh, which is a required dependency for the skill.