opensandbox-group/opensandbox ? reverse-engineered prompt
Reverse engineered prompt
Build me a secure sandbox service for AI agents that can run code, execute shell commands, manage files, and optionally open a browser or desktop inside an isolated environment.
I want it to work locally with Docker first, then also run on Kubernetes with the same basic API so I can use it in a small setup or at scale. Please include a simple command line tool and a clean API for creating a sandbox, running commands, writing and reading files, and shutting it down.
It should be fast, reliable, and safe by default, with controls for network access and a way to use external credentials without exposing them to the sandbox itself. If there are examples for common agent workflows like coding assistants, browser automation, or code execution, include them too.
If you need to check current docs or best practices while building, go ahead and look them up online.
Are you gonna build this?
make sure you review the code using coderabbit