Claw Code

Claw Code

coding-agent

面向实际试用和选型:Claw Code is an open-source AI coding agent framework described as a clean-room Python and Rust rewrite of Claude Code-style agent harness architecture.

2 views0 likes0 uses

Capabilities

  • 面向实际试用和选型:Coding-agent harness with Python orchestration and Rust runtime work, positioned as a self-hostable alternative for agentic engineering experiments.
  • 面向实际试用和选型:Focuses on tools, commands, query engine, sessions, memory, permissions, MCP integration, API clients, and streaming.
  • 面向实际试用和选型:Useful for studying coding-agent architecture, not only for using a finished IDE assistant.
  • 面向实际试用和选型:Official site explains the clean-room claim, origin timeline, architecture modules, and comparison with Claude Code.
  • 面向实际试用和选型:Can help teams prototype custom agent harness behavior without depending entirely on a proprietary CLI.
  • 面向实际试用和选型:Requires legal, provenance, and dependency review because the broader ecosystem emerged around a controversial source-map leak event.

Use Cases

  • 试用场景: 面向实际试用和选型:Clone into a disposable environment and run only on a toy repository to inspect how it handles planning, tools, sessions, and permissions.
  • 试用场景: 面向实际试用和选型:Use it as an architecture study for teams building internal coding-agent runners, especially around memory and MCP integration.
  • 试用场景: 面向实际试用和选型:Compare behavior against Claude Code, Codex CLI, OpenCode, Cursor agents, and GitHub Copilot agent workflows.
  • 试用场景: 面向实际试用和选型:Ask it to implement a small refactor with tests and inspect whether the edit loop is auditable and reversible.
  • 试用场景: 面向实际试用和选型:Review lockfiles, installation scripts, and network access before allowing it near private repositories.
  • 试用场景: 面向实际试用和选型:Use the official architecture pages as a checklist for what your own coding-agent governance needs to cover.

Examples and Source Notes

  • 官网: https://claw-code.codes/
  • 官方仓库: https://github.com/ultraworkers/claw-code
  • 文档: https://claw-code.codes/
  • Logo/图片来源: GitHub organization avatar uploaded as visible logo; official Claw Code site and repository are cited.
  • 风险检查: Treat it as an experimental framework; verify provenance, dependencies, and legal posture before enterprise use.

Related Agents

Claude Code

Claude Code

coding-agent

面向实际试用和选型:Claude Code is Anthropic's terminal-native coding agent for reading repositories, editing files, running commands, using MCP tools, and carrying software tasks through a plan-edit-test loop.

200
OpenAI Codex

OpenAI Codex

coding-agent

面向实际试用和选型:OpenAI Codex is OpenAI's software-engineering agent family spanning Codex CLI, IDE, desktop, and cloud task workflows for coding, bug fixing, and code review.

200
Google Jules

Google Jules

coding-agent

面向实际试用和选型:Google Jules is Google's autonomous coding agent for asynchronous software tasks, running work in cloud virtual machines and returning verified changes.

200