claude-in-mobile

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

• Third-party content exposure detected (high risk: 0.90). This skill can ingest and interpret arbitrary third-party content displayed on the device (e.g., via open-url which opens arbitrary URLs, screenshots/annotate/analyze-screen and ui-dump which capture and parse on-screen web/social content, plus get-clipboard/pull-file/logs that read user-provided data), so the agent would read untrusted public/user-generated content as part of its workflow.