What is analysis_claude_code CLAUDE.md?

Project instructions for a research repository that studies how Claude Code version 1.0.33 is built by examining and explaining its source code. They describe the analysis folders, documents, scripts, and common work areas.

Use them when locating analysis results, processing source chunks, writing research documentation, or running the repository's analysis tools.

They give a coding agent the repository map and the intended method for working with deobfuscated and analyzed code. This reduces the chance of editing the wrong stage, report, or generated file.

analysis_claude_code CLAUDE.md — Instructions file by ThreeFish-AI

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

agentmods

npx agentmods add instructions/threefish-ai/analysis_claude_code/claude-md

Clone the repo

git clone --depth 1 https://github.com/ThreeFish-AI/analysis_claude_code

Per session 961 This file is loaded in full into every session.

When invoked 961 The same file — it is already loaded in full.

Security scan A 0 findings. Scan, not verified.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5	$0.00961	$0.00961
Opus 5	$0.00481	$0.00481
Sonnet 5	$0.00192	$0.00192
Haiku 4.5	$0.00096	$0.00096

Measured 2d ago against content hash ff35f0b7a525, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

analysis_claude_code CLAUDE.md scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

CLAUDE.md · 120 lines

How it starts

The opening of the file, as written. The whole thing — 120 lines — stays where its author put it; the contents beside it link to each section on GitHub.

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

Repository Overview

This is a research repository containing reverse engineering analysis of Claude Code v1.0.33. The repository focuses on understanding the architecture, mechanisms, and implementation patterns of modern AI Agent systems through systematic analysis of obfuscated source code.

Project Structure

Main Analysis Workspace

claude_code_v_1.0.33/stage1_analysis_workspace/ - Complete analysis results for v1.0.33
chunks/ - Deobfuscated code chunks (102 files) split from the main CLI file
docs/ - Comprehensive technical documentation and analysis reports
scripts/ - Analysis tools and utilities for processing obfuscated code

Key Directories

analysis_results/merged-chunks/ - Optimized and merged code blocks
docs/Open-Claude-Code/ - Open source reconstruction project template
work_doc_for_this/ - Project methodology and standard operating procedures

Common Commands

Based on the analysis workspace structure, the following commands are commonly used:

Code Analysis and Processing

# Beautify and format obfuscated code
node scripts/beautify.js source/cli.mjs

# Split large files into manageable chunks
node scripts/split.js cli.beautify.mjs

# Merge and improve code chunks
node scripts/merge-again.js

# LLM-assisted analysis of code segments
node scripts/llm.js

Development Commands (Open-Claude-Code Project)

# Build TypeScript project
npm run build

# Run tests with coverage
npm run test:coverage

# Performance benchmarking
npm run benchmark

# Code quality checks
npm run lint
npm run validate

# Development mode
npm run dev

Architecture Insights

Core Technical Discoveries

h2A Async Message Queue System - Real-time steering mechanism with dual-buffer architecture
Multi-layered Agent Architecture - Main agent loop (nO), sub-agents (I2A), and task-specific agents
Intelligent Context Management - 92% threshold auto-compression with wU2 compressor
6-Layer Security Framework - From UI validation to execution sandboxing

Read the full file on GitHub · 120 lines

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

2d ago First seen · 120 lines · 961 tokens per session scan A ff35f0b7a525

Subscribe to this mod's changes

analysis_claude_code CLAUDE.md is an instructions file published in the GitHub repository ThreeFish-AI/analysis_claude_code (312 stars, last pushed 10mo ago), licensed Apache-2.0. It adds 961 tokens to every session, about $0.0048 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.

Research

vscode buildNext.instructions.md

Working notes and architecture documentation for the new esbuild-based build system in build/next. Use when making changes to the new build pipeline (transpile/bundle commands, NLS plugin, source-map handling, resource copying, or self-hosting watch tasks).

microsoft/vscode · 6,785 tokens

analysis_claude_code CLAUDE.md