smartlink-ip-workbench

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

• Third-party content exposure detected (low risk: 0.10). Skill runtime path is editing/serving a user’s local HTML and calling a local MCP proxy endpoint (e.g., /api/competitor-patent-search) with user-provided competitor names; it does not require the agent to ingest arbitrary outsider-authored free text from monitored queues/feeds/tickets without first selecting specific provider items.