captcha

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

• Third-party content exposure detected (low risk: 0.10). The workflow involves browsing and interacting with web pages where the agent navigates to arbitrary URLs chosen by the user, making page content and image grids accessible to the agent and vision models at runtime.

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

• Potentially malicious external URL detected (medium risk: 0.30). The skill makes runtime HTTP requests to third-party CAPTCHA solving services using hardcoded domain URLs (2captcha.com, capmonster.cloud, anti-captcha.com), which constitute third-party operational dependencies with common/reputable unofficial provenance.