outlook

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

• Third-party content exposure detected (high risk: 0.85). Outlook actions (e.g., outlook_list_messages, outlook_get_message, outlook_search_messages, outlook_query_emails) necessarily ingest email body text authored by the email sender (outsider) into the agent’s runtime context via the ClawLink CLI/API responses.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs running "npx @useclawlink/cli" (package: https://www.npmjs.com/package/@useclawlink/cli), which fetches and executes remote npm package code at runtime, so the external dependency executes remote code and can directly control agent behavior.