rpoisel

That's very helpful. Thanks for sharing this!

I am continuously working on my AI sandbox framework: Agent Circus.

Lately it supports running agents in Podman containers. It's also possible to independently update Agent harness versions without updating the CLI tool.

One of the next items on the roadmap is tidying up the architecture to clearly define what's in the core of that framework and what's plugged into the framework on top of it - kind of a plugin system.

Plugins may bundle settings such as what commands need to be run when building the container images, startup commands that need to run when starting the containers or the definition of information that should be forwarded from the host into agents' containers (environment variables, networking settings, etc.).

So far, I am hoping for (more) external contributions as well as someone who is willing to share ideas or talk about the concepts around that framework. :-)

https://github.com/Embedded-Focus/agent-circus

Don't want to say it's better, but I implemented Agent Circus (https://github.com/Embedded-Focus/agent-circus) which allows to lock AI agent harnesses into docker containers.

I'm using it as my main driver since months.

Support for running agent harnesses in unprivileged podman containers is on my feature list. :-)

Thanks for the hint! That's exactly what I was looking for as I have never looked into raycasting but having an interest in it.

Direct Links:

- https://nthorn.com/articles/batch_raycaster/ (batch variant)

- https://lodev.org/cgtutor/raycasting.html (general intro)

I have tried various git clients and this one is by far the best. Bold statement, but I think it's worth learning Emacs just to be able to use it. :-)

I think we have to take that literally: They only translate C code to Rust. Not C++.