Now with AI I wonder if it’s possible to just let agents build a perfect emulation of the game. It reminds me of fuzzers. You let the agent go loose on the game and it brute forces every possible state. Then recreates the code. It’s very inefficient- but it probably works.
You’re going to brute force every possible state of a sandbox building game. See you on the other side of the heat death of the universe; hope you stocked up on Claude Code credits.
I dunno. I just let Claude build a python script that calls Claude code though subprocess.run().
I recently made a sort of Autoresearch with that approach. The script calls Claude Code to create a hyphotesis, then code based on that, evaluate- rinse and repeat. I am still trying to figure out if I am actually on to something or just burning tokens. Jury is still out.
That's totally a valid approach! Especially for a very specific workflow you are looking for. For the cases I cover in cook, I had done those patterns enough times that I figured it was time to build a tool/skill for Claude so that I didn't have to explain it as much and also not have to wait for claude to code it up, and possibly interpret me wrong. Now ask claude to "/cook race 3 of foo plan with review, pick the best" and it knows what to do.
I think you're onto something, but I would add that it's sort of like a live REPL that has an integrated agent but with extra steps.
I haven't used python much but I wouldn't be surprised if you can set up a sufficiently powerful REPL with it. I know Julia can do it very well and it's a very similar language. Obviously there are powerful Lisps that do this very well as well.
Hey scrappyejoe, way looks pretty cool. The goal of cook is to be unopinionated, exposing primitives for the shape of workflows as opposed to defining what happens in those workflows. Cook is something that way could use under the hood.
Has anyone tried to set up a container and let prompt Claude to escape and se what happens? And maybe set some sort of autoresearch thing to help it not get stuck in a loop.
I want this the other way around as a plugin in Chrome. So I can easily spot the bullshit. And I would prefer that it translated LinkedIn to something salty, maybe pirate speak?
I use CLAUDE.md to describe the project. I use Claude to help write the spec. Then I let it run. When the context gets too crammed, I have it build SKILLS.md. I’ll probably have it rewrite CLAUDE.md after a while.
Then it will write tests, deployment scripts, commit messages. Yeah, everything.
Maybe someone knows law can answer this.
Is it a crime to ”unredact” files in the US? You probably know that the information is classified since you are putting in the work.
Where I live I believe it’s a crime if you share information that is classified even if it’s leaked.
So I would not publicly brag about this online.
In the US this is protected by the first amendment. Exceptions apply only for military and government employees who agree to prosecution in such cases as a condition for employment or enlistment (getting a clearance, basically). For everyone else it is lawful.
reply