Maturana aims to run Codex-style agents with stronger isolation
Maturana is an tool for running AI agents with stronger security controls. On Linux, it uses and KVM, and on Windows, it uses Hyper-V, so agents run inside hardware-level isolation instead of only a container. Pipelock keeps sensitive data such as API keys and credit card numbers away from the agent, while an egress proxy limits which systems the agent can contact.
The tool is built for Codex workflows, where prompts can create agents, tools, and skills, but Codex is not required because the can also be used. It also includes a text interface and a for managing agents. An internal WASM engine lets agents create their own tools when needed, but keeps them inside a sandbox.
Instead of using markdown files for shared knowledge, it includes a built-in , and the core is written in Rust.
Key points
- Maturana is an tool for running AI agents with stronger security controls.
- It uses and KVM on Linux and Hyper-V on Windows for hardware-level isolation.
- Pipelock is meant to keep API keys and credit card numbers out of the agent’s reach.
- An egress proxy controls which outside systems an agent can contact.
- It is designed around Codex workflows, but it can also be used from the .