{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/35536efaa21b4251b197a3bcbef82763\" frameborder=\"0\" width=\"1110\" height=\"832\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":832,"width":1110,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":832,"thumbnail_width":1110,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/35536efaa21b4251b197a3bcbef82763-f41b0727ae771649.gif","duration":257.262,"title":"EVE Security Protects Agentic AI Interactions - Cursor","description":"This Loom demonstrates how EVE Security protects agentic AI interactions across productivity tools and enterprise systems. Sharon shows Cloud and Cursor agents communicating with Databricks and Linear, with EVE Guard enforcing policies visible in the Explorer graph view. She explains global and system specific rules, including read only access to Linear for Cursor and blocking transmission of PII or HPI, while Gmail interaction has no policy. In the demo, Cursor successfully reads a Linear ticket but is blocked when attempting to change ticket status to open, and additional attempts to fetch forbidden data like a social security number are intercepted and blocked."}