claw-code - An open-source Claude Code source code rewrite project

What is claw-code

claw-code is aClaude Code Clean-room Rewrite project launched after the source code leak. To avoid legal risks, the author used OpenAI Codex's OmX workflow orchestration to port the core architecture from TypeScript to a Python implementation in just a few hours. The project retains the original Agent workflow and tool orchestration capabilities as an open-source alternative, and is currently being further ported to Rust to provide a more efficient runtime. The project emphasizes "Better Harness Tools" to build a truly usable open-source Agent development framework.

Main features of claw-code

• Agent workflow orchestration: The core architecture of the Agent system responsible for tool connection, task orchestration, and runtime context management.
• Tool system: A modular tool framework that provides tool call chain management and tool port metadata definition.
• Command management: A modular command system and port metadata management that encapsulates various development operations.
• Task scheduling: A task state management and scheduling system that supports asynchronous processing and execution loops.
• Query engine: A query processing engine that supports code understanding, context retrieval, and codebase analysis.
• CLI interactive interface: Provides a command-line entry point for viewing migration summaries, inventories, and interactive operations.
• Runtime context management: A persistent context management system that maintains conversation history and execution state.
• Automated code generation: Automated development capabilities for code understanding, generation, and refactoring based on AI.

How to use claw-code

• Environment preparation: Ensure that Python 3 is installed locally, clone the project repository locally, and then enter the root directory.
• View overview: By running the main module command, you can view summary information about the current Python porting work as well as the complete workspace structure inventory.
• Explore modules: Use the corresponding command to list system submodules, available command interfaces, and supported tools.
• Run tests: Run the test discovery command to execute the verification test suite and ensure the correctness of the ported code.
• Comparison audit: When the original archive exists locally, you can run a consistency audit to check the degree of functional alignment between the ported version and the original code.

Key information and usage requirements of claw-code

• Project nature: A Clean-room Rewrite after the Claude Code source code leak, not an official archive.
• Tech stack: Currently a Python implementation (src/ directory), the Rust version (dev/rust branch) is about to be merged, pursuing memory safety and high performance.
• Core positioning: An Agent development framework that provides Harness capabilities such as tool orchestration, command management, task scheduling, and query engine.
• Development method: Uses oh-my-codex (OmX) AI workflow orchestration throughout,teammode parallel review, ralph mode persistent verification.
• Current status: The basic Python framework is already usable, but it has not yet achieved full runtime equivalence with the original TypeScript system; the Rust version will be the final form.
• Environment: Python 3 is required; clone the repository and run it.

Core advantages of claw-code

• Legal safety: It adopts a clean-room rewrite approach and is independently implemented from scratch, completely avoiding the copyright infringement and legal litigation risks caused by using leaked source code.
• Open source freedom: As a fully open source project, it is not subject to commercial software restrictions, and developers can freely modify, extend, and deeply customize the Agent workflow and toolchain.
• Technical advancement: It is being ported to the Rust language, which provides memory safety, higher runtime performance, and system-level stability compared with the original TypeScript implementation.
• AI-native development: The entire development process uses the oh-my-codex AI workflow for orchestration, representing the engineering paradigm and automated build capabilities of future AI-assisted programming.
• Functional completeness: It fully preserves Claude Code's core Harness capabilities, including key functions such as Agent workflow orchestration, the tool system, command management, and task scheduling.

claw-code project address

• GitHub repository：https://github.com/instructkr/claw-code

Comparison of claw-code with similar competitors

Application scenarios of claw-code

• AI-assisted programming: As an open-source alternative to Claude Code, it provides intelligent programming assistant capabilities for automatic code generation, completion, refactoring, and error fixing.
• Intelligent codebase analysis: Uses a query engine to deeply understand the structure of large codebases, enabling cross-file semantic retrieval, architecture analysis, and automatic documentation generation.
• Automated workflow orchestration: Uses the Harness system to orchestrate complex multi-step development tasks, such as automated builds, deployment pipelines, and continuous integration script management.
• Toolchain integration and extension: As a modular Agent framework, it integrates various development tools (Git, testing frameworks, Linters) to achieve unified command scheduling and tool invocation.
• Code migration and refactoring: Supports automated refactoring of large-scale codebases, language migration (such as Python to Rust), and architectural modernization.