{"type":"video","version":"1.0","html":"<iframe src=\"https://www.loom.com/embed/8d10d5f2f2384133b5b0f63897fb098f\" frameborder=\"0\" width=\"1920\" height=\"1440\" webkitallowfullscreen mozallowfullscreen allowfullscreen></iframe>","height":1440,"width":1920,"provider_name":"Loom","provider_url":"https://www.loom.com","thumbnail_height":1440,"thumbnail_width":1920,"thumbnail_url":"https://cdn.loom.com/sessions/thumbnails/8d10d5f2f2384133b5b0f63897fb098f-bee68e6e1f2ae96b.gif","duration":188.499,"title":"Sudound AI Agent Firewall Live Demo 🚀","description":"Hi, I’m Evgeny, and this is a demo of Sudound running as a local Solana RPC sidecar, with no faked data. On the left pane you can see real time Solana blocks, the middle pane shows stats like memory and response time, and the bottom pane shows firewall logs. I launch multiple AI agent scenarios on the right, including a normal transfer allowed, an interrogate flow that blocks and self corrects, an EVM sandwich attack blocked for large slippage, and a Trojan SBF style token swap blocked with hidden routing. I also show a Monte Carlo simulation for revenue generation that must run within block time, and I request you pause the video to view scenario details."}