verify-keyboard-access

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

• Third-party content exposure detected (low risk: 0.10). The skill instructs the agent to browse and test arbitrary web pages and run UI automation tools on target web applications where the page content may contain untrusted outsider text.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill instructs fetching external capability and documentation files via curl from docs.reticle.sh at runtime, introducing unverified remote dependencies.