Summary

This repository provides a method to decompile and study the source code of cloud-code, a tool likely related to AI-assisted development or code analysis. It targets developers seeking to understand the internal architecture and implementation of such proprietary software, enabling local exploration and compilation.

How It Works

The project reconstructs the original TypeScript/TSX source code by parsing the source map (cli.js.map) of a specific npm package version (@anthropic-ai/claude-code@2.1.88). A provided Node.js script extracts source file content (sourcesContent) from the map, effectively de-obfuscating and restoring the complete codebase for local analysis. The build is adapted for the Bun runtime, with specific optimizations for Apple Silicon.

Quick Start & Requirements

• Primary Install/Run: The README details a multi-step restoration process:
  1. npm pack @anthropic-ai/claude-code --registry https://registry.npmjs.org
  2. tar xzf anthropic-ai-claude-code-2.1.88.tgz
  3. Execute a provided Node.js script to parse package/cli.js.map and reconstruct source files into ./claude-code-source.
• Prerequisites: Bun runtime, Node.js (for the extraction script).
• Hardware: Optimized for Apple Silicon (M1/M2/M3). Other architectures require manual adaptation.
• Links: Experience Demo, Study Guide.

Highlighted Details

• Restores 4,756 source files from the cloud-code package via source map data.
• The core source code comprises 1,906 files across src/ (1902 files) and vendor/ (4 files).
• Detailed module breakdown includes extensive utils/ (564 files), components/ (389 files), commands/ (207 files), and tools/ (184 files).
• Features a custom Ink-based terminal UI engine and native modules for audio capture and image processing.

Maintenance & Community

No specific details regarding maintainers, community channels (e.g., Discord, Slack), or project roadmaps are provided in the README.

Licensing & Compatibility

The README does not specify a license. The project is described as a "cracked version" ("破解版") for "source code study" ("源码学习"), indicating it's a decompiled or reverse-engineered artifact of potentially proprietary software. Commercial use or integration into closed-source projects is highly discouraged and likely prohibited.

Limitations & Caveats

This project is presented as a decompiled "cracked version" for educational purposes, raising significant legal and ethical concerns regarding the original software's intellectual property. Installation and runtime functionality for CPU architectures other than Apple Silicon require user-driven adaptation. The completeness and stability of the restored codebase are not guaranteed.