Back to News

External sandbox pattern for coding agents with three server implementations

#coding agents#sandboxing#security#webassembly

Simon Willison and Prime Radiant introduced the external sandbox pattern for coding agents, where agent-executed code runs in a dedicated sandbox separate from the agent itself. Three sandbox server implementations were released: external-sandbox-v86 (Node.js with v86 running Alpine Linux in WebAssembly), external-sandbox-container (Python with Alpine Linux in Apple Containers), and external-sandbox-quicksand (Python using Microsoft's Quicksand library).

Coverage timeline

  1. Simon Willison

    External sandboxes for coding agents I've been exploring new ways to sandbox code that is executed by agents, with a focus on making it safer to try out local models without worrying that a mistake or prompt injection attack might cause them to take damaging actions on a machine. I worked on this project with Prime Radiant , the AI research lab behind Superpowers and other projects . My post on the Prime Radiant blog introduces the external sandbox pattern, where code executed by the agent runs inside a dedicated sandbox while the agent itself lives outside of that sandbox. To demonstrate this pattern we've released three sandbox server implementations: external-sandbox-v86 runs a Node.js server which then uses v86 to run Alpine Linux inside WebAssembly. external-sandbox-container is a Python web server which runs the sandbox in Alpine Linux inside an Apple Containers instance. external-sandbox-quicksand is another Python web server, this time using Microsoft's Quicksand sandboxing lib

News updates every hour. Subscribe in your RSS reader to get new stories automatically.

Subscribe