Claude Code Unpacked : A visual guide

A 500k line codebase for an agent CLI proves one thing: making a probabilistic LLM behave deterministically is a massive state-management nightmare. Right now, they're great for prompting simple sites/platforms but they break at large enterprise repos.

If you don't have a rigid, external state machine governing the workflow, you have to brute-force reliability. That codebase bloat is likely 90% defensive programming; frustration regexes, context sanitizers, tool-retry loops, and state rollbacks just to stop the agent from drifting or silently breaking things.

The visual map is great, but from an architectural perspective, we're still herding cats with massive code volume instead of actually governing the agents at the system level.

Author here. I built this in a few hours after the Claude Code leak.

I've been working on my own coding agent setup for a while. I mostly use pi [0] because it's minimal and easy to extend. When the leak happened, I wanted to study how Anthropic structured things: the tool system, how the agent loop flows, A 500K line codebase is a lot to navigate, so I mapped it visually to give myself a quick reference I could come back to while adapting ideas into my own harness and workflow.

I'm actively updating the site based on feedback from this thread. If anything looks off, or you find something I missed, lmk.

[0] https://pi.dev/

I built a site that lets you explore and browse all the Claude Code prompts in a structured way:

https://ccprompts.info

I know it seems counter-intuitive but are there any agent harnesses that aren’t written with AI? All these half a million LoC codebases seem insane to me when I run my business on a full-stack web application that’s like 50k lines of code and my MvP was like 10k. These are just TUIs that call a model endpoint with some shell-out commands. These things have only been around in time measured in months, half a million LoC is crazy to me.

> 500k lines of code

Isn't it a simple REPL with some tools and integrations, written in a very high level language? How the hell is it so big? Is it because it's vibecoded and LLMs strive for bloat, or is it meaningful complexity?

If it was 2020, it would be hard to imagine that after some hours/days you getting a visual representation of the leak with such detailed stats lol

Dang! Glad to see others doing this. I totally made this site yesterday like 11 hours ago :/ but did not get the traction.

I love your implementation.

Here was my first stab:

https://news.ycombinator.com/item?id=47595140

https://brandonrc.github.io/journey-through-claude-code/

Same guide for opencode would be nice too

Here's a codeberg repo with the leaked source: https://codeberg.org/wklm/claude-code

Even today, I'm still astounded that there are people capable of building a gorgeous and interesting site like this in less than 2 days...

I don't know why people obsess and spend so much time on this codebase. It isn't (and never was)alien technology. It's just mediocre typescript generated by an LLM

I guess they really do eat their own dogfood and vibe code their way through it without care for technical debt? In a way, it’s a good challenge, but it’s fairly painful to watch the current state of the project (which is about a year old now, so it should be in prime shape).

Feel free to add this to Awesome Claude code. https://github.com/rosaboyle/awesome-cc-oss

Appreciate the effort, but this is very basic and nothing you need the source code to understand. I was expecting a deep dive into what specific decisions they made, but not how an loop of tool calls works

Thanks to Claude Code, we got such a beautifully polished and dazzling website that gives a complete introduction to itself the very moment the leak happened :)

Nice job - I'm a fan. Makes it easy to get the big picture so I know where to dive in.

> also related: https://www.ccleaks.com

This deployment is temporarily paused

Okay those "hidden features" are amazing, especially the cross-session referencing. I hope we can look forward to that in the future

Also I definitely want a Claude Code spirit animal

ccleaks.com seems to be "temporarily paused" from Vercel.

Here is another one that goes in depth as well: www.markdown.engineering for anyone going deep on learning.

Kairos and auto-dream are more interesting than anything in the agent loop section. Memory consolidation between sessions is the actual unsolved problem. The rest is just plumbing tbh

There's this weird thing about AI generated content where it has the perfect presentation but conveys very little.

For example the whole animation on this website, what does it say beyond that you make a request to backend and get a response that may have some tool call?

Btw, the 500K is just the source - it does not include tests. I would imagine there are at least 2-4x tests.

A year ago I wouldn't have guessed a TUI could be a competitive advantage. But "harness engineering" became a thing, and it turns out the agent wrapper — tool orchestration, context management, permission flows — is where real product value lives. Not as much as the models themselves, but more than most people expected. This leak is a painful reminder of that.

Looks like ccleaks is down eek - not long before ccunpacked has same fate.

Is it just me or do I not find the Claude Code application that fascinating?

I use it all day and love it. Don't get me wrong. But it's a terminal-based app that talks to an LLM and calls local functions. Ooookay…

I'm developing an agent focused on A2A, support for small models, and privacy (https://swival.dev).

I looked at the leaked code expecting some "secret sauce", but honestly didn't found anything interesting.

I don't get the hype around Claude Code. There's nothing new or unique. The real strength are the models.

would be nice if the transformers code for one of these frontier LLM models got leaked, HN will have a field day with a reveal like that

Nice presentation. The reality is there is nothing really special about the claude code harness?

Really nice visualisation of this, makes understanding the flow at a high levle pretty clear. Also the tool system and command catalog, particularly the gated ones are super interesting.

No mention of undercover mode?

However, excellent development practices involve modularizing code based on functional domains or responsibilities.

The utils directory should only contain truly generic, business-agnostic utilities (such as date retrieval, simple string manipulation, etc.).

We can see that the code produced by Vibe is not what a professional engineer would write. This may be due to the engineers using the Vibe tool.

Nice site. I might suggest moving SendMessage to the Hidden Features as they don't appear to have implemented a ReadMessage or ListMessages tools.

Ah, good well-architected code, finally... With most of the code in utils/other :D

I hope /Buddy is ported across to OpenCode.

So it does use ripgrep and not unix grep. [0] I knew it from some other commenters here on HN, but it's nice to see it in the source as well.

0 - https://github.com/zackautocracy/claude-code/blob/main/src/u...

This is AI slop.

First command I looked at:

  /stickers:
  
  Displays earned achievement stickers for milestones like first commit, 100 tool calls, or marathon sessions. Stickers are stored in the user profile and rendered as ASCII art in the terminal.

What is the motivation for someone to put out junk like this?

I prefer this mapping from Nikita @ CosmoGraph: https://run.cosmograph.app/public/dfb673fc-bdb9-4713-a6d6-20...

why do people care so much? it's just an agentic loop

its April fools joke. this has really gone wide

519K lines of code for something that is using the baseline *nix tools for pretty much everything important, how do they even manage to bloat it this much? I mean I know how technically, but it's still depressing. Can't they ask CC to make it good, instead of asking it to make it bigger?

I mean, I get it: vibe-coded software deserves vibe-coded coverage. But I would at least appreciate it if the main part of it, the animation, went at a speed that at least makes it possible to follow along and didn't glitch out with elements randomly disappearing in Firefox...

How is this on the front page?

nice example: Find all TODO spin the AI machine

i do shift ctrl F

I think this is unethical, and "everyone else is also doing it" is not a valid excuse.

I expect dozens more "research articles" that

- find nothing - still manage to fill entire lages - somehow have a similar structure - are boring as fuck

At least this one is 3/4, the previous one had BINGO.

Is that safe to use?

How the hell is it 500k lines?

Ccleaks is down?

Source leak or free code review? I can say that there is no bad publicity.

what is so fascinating about claude code. we have codex that is open source already. is there something special to learn from claude code?

this claude code leak is such a fuck up...

The fact that now every agent designer knows what was already built is a huge shot of steroids to their codebase!