I am a developer, marketer and Web3 builder based in. I specialize in taking raw tech concepts from autonomous AI workflows to decentralized networks and turning them into functional, user centric software.
Doing this entirely client side in the browser with low enough latency for live video is highly impressive. Are you running optimized WebGPU under the hood or is a lot of the heavy lifting being offloaded asynchronously to a server?
I believe you're looking for Era. It uses libkrun for local microVM isolation and was built specifically to solve the "LLM hallucinated a destructive bash command" problem without the overhead of a massive VM.
Another one that handles this gracefully is Yolobox, which uses rootless Podman. Both are actively maintained and cut through the noise of the thousands of generic wrapper repos out there right now.