agents

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

• Third-party content exposure detected (high risk: 0.80). The skill explicitly lets agents call webhook and client tools with arbitrary api_schema URLs and client-side fetches (see references/client-tools.md and SKILL.md tools examples), and the webhook/client responses are returned to and acted on by the agent, so untrusted third-party content can influence agent behavior.

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

• Potentially malicious external URL detected (high risk: 0.70). The embed script https://unpkg.com/@elevenlabs/convai-widget-embed is loaded at runtime via a and therefore fetches and executes remote JavaScript code the skill relies on for the widget (remote code execution at runtime).