Skip to content

Claude Code Rust

Rust 全量重构的 Claude Code - 性能提升 2.5x,体积减少 97% | High-performance Rust implementation of Claude Code with 2.5x faster startup and 97% smaller binary

Install

$npx tauri dev
Claude Code
claude mcp add claude-code-rust -- npx tauri dev
Cursor / Claude Desktop (mcp.json)
{
  "mcpServers": {
    "claude-code-rust": {
      "command": "npx",
      "args": [
        "tauri",
        "dev"
      ]
    }
  }
}

Install command taken from the project's README or registry entry. Always review code before running it.

Claude Desktop (Tauri Edition)

A desktop client for AI chat assistants, built with Tauri 2.0, Rust, and React. Provides a native-feeling GUI for interacting with Claude and other LLM providers, with local data persistence and a plugin-style architecture for extending functionality.


What This Is

This is a personal/community desktop application that wraps AI chat APIs in a native window. It is not an official Anthropic product, and it is not a polished commercial release. It is under active development: features change frequently, some things are incomplete, and you should expect rough edges.

The app uses Tauri 2.0's Rust backend to handle file system access, process management, SQLite storage, and HTTP proxying, while the React 19 frontend provides the chat UI, settings panels, and auxiliary views.


Features

Chat and Models

  • Multi-provider chat with streaming responses (SSE)
  • Multiple model support across Anthropic, OpenAI, DeepSeek, and other compatible APIs
  • Chat history persistence in SQLite
  • Context-aware memory system that surfaces relevant past conversations
  • Markdown rendering with syntax highlighting, KaTeX math, and Mermaid diagrams
  • Voice input support

Tooling and Extensions

  • MCP (Model Context Protocol) server integration -- connect external tools that the AI can invoke
  • Multi-agent orchestration -- coordinate multiple AI agents working on subtasks in parallel
  • Skills system -- user-extensible plugin modules for specialized capabilities
  • Slash commands for quick actions within the chat input
  • Built-in terminal panel (xterm.js)
  • Code diff viewer for reviewing AI-generated code changes
  • Code execution panel for running code snippets

Panels and Views

  • File explorer for browsing the local filesystem
  • Knowledge base panel for managing reference documents
  • Research panel for multi-step web research workflows
  • Artifact preview for rendered HTML/JSX/WebGPU outputs
  • Live preview panel for web content
  • Document panel with DOCX and PDF preview
  • Swarm collaboration view for multi-agent task management
  • IM integration panel (Feishu/WeChat)
  • Analytics and cost tracking dashboard
  • GitHub integration panel
  • App Studio for building mini-apps within the desktop

Desktop Integration

  • System tray icon
  • Native file dialogs
  • Clipboard integration
  • Desktop notifications
  • Automatic updater (checks for new releases)

Permissions

  • Configurable permission modes: ask every time, auto-accept edits, plan-only, or full bypass
  • Per-tool permission controls

Tech Stack

Frontend

Technology Purpose
React 19 UI framework
TypeScript 5.7 Type system
Vite 6 Build tooling
Tailwind CSS 3 Utility-first styling
Zustand 5 State management
React Router 6 Client-side routing
xterm.js 5 Terminal emulation
KaTeX Math rendering
Mermaid Diagram rendering
recharts Charting
highlight.js / react-syntax-highlighter Code highlighting
react-markdown / remark / rehype Markdown pipeline

Similar plugs

The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.

Fair-code workflow automation platform with native AI capabilities. Combine visual building with custom code, self-host or cloud, 400+ integrations.

Java 面试 & 后端通用面试指南,覆盖计算机基础、数据库、分布式、高并发、系统设计与 AI 应用开发

Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype…

Explore related