How to code Claude Code in 200 lines of code • Hacker News Board

Re: How to code Claude Code in 200 lines of code

#91

Post by matsemann » Thu, Jan 08, 2026, 10:01 PM UTC

NitpickLawyer wrote: post #83

FrontierProject wrote: post #71

Earlier quoted context omitted.

This site has gone full Tower of Babel. I've seen at least a thousand "AI comment" callouts on this site in the last month and at this point I'm pretty sure 99% of them are wrong. In fact, can someone link me to a disputed comment that the consensus ends up being it's actually AI? I don't think I've seen one.

You know how the chicken sexers do their thing, but can't explain it? Like they can't write a list of things they check for. And when they want to train new people they have them watch (apprentice style) the current ones, and eventually they also become good at doing it themselves? It's basically that. I can't explain it (I tried listing the tells in a comment below), but it's not just a list of things you notice. Yo…

> You know how the chicken sexers

That's certainly a novel and confusing entry in my search history.

Re: How to code Claude Code in 200 lines of code

#92

Post by jake-coworker » Thu, Jan 08, 2026, 10:02 PM UTC

bredren wrote: post #80

libraryofbabel wrote: post #35

It's a great point and everyone should know it: the core of a coding agent is really simple, it's a loop with tool calling. Having said that, I think if you're going to write an article like this and call it "The Emperor Has No Clothes: How to Code Claude Code in 200 Lines of Code", you should at least include a reference to Thorsten Ball's excellent article from wayyy back in April 2025 entitled "How to Build an Age…

I've been exploring the internals of Claude Code and Codex via the transcripts they generate locally (these serve as the only record of your interactions with the products)[1]. Given the stance of the article, just the transcript formats reveals what might be a surprisingly complex system once you dig in. For Claude Code, beyond the basic user/assistant loop, there's uuid/parentUuid threading for conversation chains,…

IMO these articles are akin to "Twitter in 200 lines of code!" and "Why does Uber need 1000 engineers?" type articles.

They're cool demos/POCs of real-world things, (and indeed are informative to people who haven't built AI tools). The very first version of Claude Code probably even looked a lot like this 200 line loop, but things have evolved significantly from there

Re: How to code Claude Code in 200 lines of code

#93

Post by johnsmith1840 » Thu, Jan 08, 2026, 10:05 PM UTC

Here's the bigger question. Why would you?

Claude code feels like the first commodity agent. In theory its simple but in practice you'll have to maintain a ton of random crap you get no value in maintaining.

My guess is eventually all "agents" will be wipped out by claude code or something equivalent.

Maybe not the companies will die but that all those startups will just be hooking up a generic agent wrapper and let it do its thing directly. My bet is that that the company that would win this is the one with the most training data to tune their agent to use their harness correctly.

Re: How to code Claude Code in 200 lines of code

#94

Post by libraryofbabel » Thu, Jan 08, 2026, 10:09 PM UTC

aszen wrote: post #86

libraryofbabel wrote: post #59

Earlier quoted context omitted.

Yeah, there is definitely some RLVR training going on for the Claude LLMs to get them good at some of the specific tool calls used in Claude Code, I expect. Having said that, the string replacement tool schema for file edits is not very complicated at all (you can see it in the tool call schema Claude Code sends to the LLM), so you could easily use that in your own 200-300 line agent if you wanted to make sure you're…

Yeah that's one example, but I suspect they train the model on entire sequences of tool calls, so unless you prompt the model exactly as them you won't get the same results. There's a reason they won the agent race, their models are trained to use their own tools.

Agree, the RLVR tasks are probably long series of tool calls at this point doing complex tasks in some simulated dev environment.

That said, I think it's hard to say how much of a difference it really makes in terms of making Claude Code specifically better than other coding agents using the same LLM (versus just making the LLM better for all coding agents using roughly similar tools). There is probably some difference, but you'd need to run a lot of benchmarks to find out.

Re: How to code Claude Code in 200 lines of code

#95

Post by handfuloflight » Thu, Jan 08, 2026, 10:15 PM UTC

Johnny_Bonk wrote: post #87

bredren wrote: post #80

Earlier quoted context omitted.

I've been exploring the internals of Claude Code and Codex via the transcripts they generate locally (these serve as the only record of your interactions with the products)[1]. Given the stance of the article, just the transcript formats reveals what might be a surprisingly complex system once you dig in. For Claude Code, beyond the basic user/assistant loop, there's uuid/parentUuid threading for conversation chains,…

This is very interesting, especially if you could then use an llm across that search to figure out what has and maybe has not been completed, and then reinject those findings into a new Claude code session

For that you'd be better off having the LLM write TODO stubs in the codebase and search for that. In fact, most of the recent models just do this, even without prompting.

Re: How to code Claude Code in 200 lines of code

#96

Post by tomhow » Thu, Jan 08, 2026, 10:18 PM UTC

dana321 wrote: post #43

[flagged]

Please don't fulminate on HN. The guidelines make it clear we're trying for something better here. https://news.ycombinator.com/newsguidelines.html

Re: How to code Claude Code in 200 lines of code

#97

Post by mudkipdev » Thu, Jan 08, 2026, 10:19 PM UTC

Is this not using constrained decoding? You should pass the tool schemas to the "tools" parameter to make sure all tool calls are valid

Re: How to code Claude Code in 200 lines of code

#98

Post by btown » Thu, Jan 08, 2026, 10:19 PM UTC

lmeyerov wrote: post #63

Something I would add is planning. A big "aha" for effective use of these tools is realizing they run on dynamic TODO lists. Ex: Plan mode is basically bootstrapping how that TODO list gets seeded and how todos ground themselves when they get reached, and user interactions are how you realign the todo lists. The todolist is subtle but was a big shift in coding tools, and many seem to be surprised when we discuss it -…

I'm unsure of its accuracy/provenance/outdatedness, but this purportedly extracted system prompt for Claude Code provides a lot more detail about TODO iteration and how powerful it can be:

https://gist.github.com/wong2/e0f34aac66caf890a332f7b6f9e2ba...

I find it fascinating that while in theory one could just append these as reasoning tokens to the context, and trust the attention algorithm to find the most recent TODO list and attend actively to it... in practice, creating explicit tools that essentially do a single-key storage are far more effective and predictable. It makes me wonder how much other low-hanging fruit there is with tool creation for storing language that requires emphasis and structure.

Re: How to code Claude Code in 200 lines of code

#99

Post by bredren » Thu, Jan 08, 2026, 10:24 PM UTC

Johnny_Bonk wrote: post #87

bredren wrote: post #80

Earlier quoted context omitted.

I've been exploring the internals of Claude Code and Codex via the transcripts they generate locally (these serve as the only record of your interactions with the products)[1]. Given the stance of the article, just the transcript formats reveals what might be a surprisingly complex system once you dig in. For Claude Code, beyond the basic user/assistant loop, there's uuid/parentUuid threading for conversation chains,…

This is very interesting, especially if you could then use an llm across that search to figure out what has and maybe has not been completed, and then reinject those findings into a new Claude code session

I haven't written the entry yet but it is pretty incredible what you can get when letting a frontier model RAG your complete CLI convo history.

You can find out not just what you did and did not do but why. It is possible to identify unexpectedly incomplete work streams, build a histogram of the times of day you get most irritated with the AI, etc.

I think it is very cool and I have a major release coming. I'd be very appreciative of any feedback.

Re: How to code Claude Code in 200 lines of code

#100

Post by floppyd » Thu, Jan 08, 2026, 10:26 PM UTC

> This is the key insight: we’re just telling the LLM “here are your tools, here’s the format to call them.” The LLM figures out when and how to use them.

This really blew my mind back then in the ancient times of 2024-ish. I remember the idea of agents just reached me and I started reading various "here I built an agent that does this" articles, and I was really frustrated at not understanding how the hell LLM "knows" how to call a tool, it's a program, but LLMs just produce text! Yes I see you are telling LLM about tools, but what's next? And then when I finally understood that there's no next, no need to do anything other than explaining — it felt pretty magical, not gonna lie.