I open sourced it too.
It has a bit to go—like automatically monitoring reasoning traces and security—but I thought this would be easier to use than trying to maintain your own skills.md collection if the server handled more of it for you.
I’d be curious if anyone thinks I should keep working on it. I’d love someone to collaborate with on it!
It’s like 5 hours of work and I’m super early with it!
The tradeoff is that the prompt is non-deterministic, but also more flexible. It can handle unknown situations. The generated code executes much faster and reliable for known situations.
If auto-learn would reduce the non-deterministic nature, i.e. chance of failing, that would speed up software development.
Even if the agent has mcp tools avaible, you can't guarantee that it would necessarily use them right? I imagine that you'd have a lot of new tools that do nearly the same things.
Also how does this scale? If the agent has like 300 tools.
