google-chat

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

• Third-party content exposure detected (medium risk: 0.30). The Google Chat skill monitors and reads chat messages from Google Chat spaces where any participant can post arbitrary free text.

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

• Potentially malicious external URL detected (high risk: 0.70). The skill uses a custom third-party cloud function domain for handling Google OAuth flows, presenting an unverifiable runtime external control-plane dependency with weak provenance.