classify

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

• Third-party content exposure detected (low risk: 0.10). The skill workflow uses classifier.dev to classify text gathered from user-provided files or search results, which represents active searching/fetching rather than monitoring an outsider-authored queue or feed.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs the agent to make HTTP POST requests to classifier.dev as an operational instruction source/API dependency, which is an unverified third-party service domain without strong first-party provenance.