🚀 Seeking Beta Testers: Oxide v0.1.0

[gorecodes]

🚀 Seeking Beta Testers: Help us stress-test the AI Automation Kernel!

Hello everyone! 🦀

I've just released the first alpha version of this lightweight, sandboxed AI automation kernel. The system is designed to run automation tasks using Lua-based skills in a secure environment, optimized for LXC containers and Proxmox environments.

We are looking for early adopters and developers to help us stress-test the system before we move toward a stable v1.0.

⚙️ Technical Requirements (AI Gateway)

This project requires an active connection to an LLM (Large Language Model) provider for semantic processing and embeddings.

• AI Proxy: You will need LiteLLM or a similar OpenAI-compatible gateway to manage your API connections (OpenAI, Vertex AI, Anthropic, etc.).
• Configuration: Ensure your gateway is correctly configured and that the system has access to the necessary API keys and endpoints.

🎯 What we need to test:

• Sandbox Security: Try to break out of the Lua environment or trigger resource exhaustion.
• Instruction Hook: Verify that infinite loops or heavy computations are correctly throttled/killed.
• Semantic Reliability: Testing the consistency of the routing and embedding logic via the AI gateway.

🛠️ How to get started:

The easiest way to help is to download the latest release and run a few example skills:

1. Download the Release: Head over to our Releases page and grab the latest tarball.
2. Setup LiteLLM: Ensure your AI gateway is running and reachable.
3. Quick Run:

   tar -xzf <release-archive>.tar.gz
   cd <release-directory>
   ./executable-name

4. Create a Skill: Try writing a simple .lua script in the skills/ folder and see how the system handles it.

🐛 How to report issues:

If something explodes (or doesn't work as expected), please:

1. Open a new issue with the bug label.
2. Include your OS/Kernel version (e.g., Ubuntu 22.04 in LXC).
3. Attach the logs from the terminal and specify your LiteLLM configuration.