youtube

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

• Third-party content exposure detected (high risk: 0.85). The skill’s runtime workflow uses ClawLink to “list/search comments/videos/threads” from YouTube, which ingests outsider-authored free text (e.g., YouTube comment bodies and other user-generated text) into the agent’s LLM context via the YouTube action results.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running "npx @useclawlink/cli" which will fetch and execute remote code from the npm package at runtime (https://www.npmjs.com/package/@useclawlink/cli), so it has a required runtime dependency that executes external code.