canary-watch

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly fetches and monitors arbitrary deployed URLs (see "What It Watches" and examples like /canary-watch https://myapp.com), reading page content, console errors, and network requests from public sites—untrusted third-party content that could influence alerts and subsequent actions.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill explicitly loads target sites at runtime (e.g., https://myapp.com and https://staging.myapp.com) to capture console errors and performance metrics—loading those pages will execute remote JavaScript in the monitoring environment, so external content can execute code during skill runtime.