facebook

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

• Third-party content exposure detected (medium risk: 0.30). The workflow monitors Facebook comments, messages, and posts via ClawLink, allowing outsiders to submit untrusted free text into the agent's ingestion path.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs execution of npm packages from an unverified third-party CLI tool (@useclawlink/cli) which introduces an unverifiable external runtime dependency tier.