A local sandbox that provides AI agents with code execution, filesystem access, and a full GUI desktop environment for 'computer use' capabilities. It exposes tools for shell commands, multi-language code interpretation, and remote desktop automation via the Model Context Protocol.
Provides a human-in-the-loop security layer for AI agents by intercepting file operations, explaining them with a local LLM, and enforcing a deterministic policy that requires user approval for risky actions.