Claude Desktop 설치 가이드
로컬 MCP 설정을 지원하는 데스크톱 클라이언트.
MCP Servers
WIP: We are building a benchmark for evaluating the optimizations for OS!
A Model Context Protocol (MCP) server providing Django project exploration resources and optional stateful shell access for LLM assistants to interact with Django projects.
Unsupervised use or misuse may lead to unintended consequences. All AI systems must remain strictly under human oversight and control. Use responsibly, with full awareness and at your own risk.
MCP server that turns natural-language prompts into Microsoft Azure architecture diagrams (PNG) using Python Diagrams + Graphviz.
MCP Client 是一个基于 Model Context Protocol 的 Node.js 客户端实现（使用 Function Calling），它允许您的应用连接到各种 MCP 服务器，并通过大语言模型（LLM）与这些服务器交互。MCP（模型上下文协议）是一个开放协议，用于标准化应用程序向 LLM 提供上下文的方式。
Core PHP implementation for the Model Context Protocol (MCP) Client
A Model Context Protocol (MCP) server implementation that provides capabilities to interact with Moodle LMS
Ollama MCP Agent allows you to use LLM models locally on your PC for free. Using Ollama's locally installed LLM models along with MCP (Model Context Protocol) additional features, you can easily...
MCP server for Polyrouter API - unified prediction market aggregator
Agent Forge 是一款专为 用户打造的桌面工具，将分散在各个目录中的 和 配置集中到一个优雅的可视化界面，告别手动翻找目录和编辑 JSON 的烦恼。
A Model Context Protocol (MCP) server for stock data retrieval
A comprehensive Model Context Protocol (MCP) server that provides access to the AlphaFold Protein Structure Database through a rich set of tools and resources for protein structure prediction...
Agent Skills
WSL workspace support for DeepSeek Harness——无缝的 WSL 工作区使用体验，无需在 WSL 之中再安装一个dsh，安装该插件后在 GUI 里直接添加 WSL 工作区即可。
The entire knowledge around the detection lifecycle, encoded into skills. More details at https://detectionskills.io/
Google Docs for markdown files — real-time multiplayer editing, comments, suggestion mode, edit history, and a CLI built for AI agents.
基于 Claude Code 官方 skill-creator，我融入了自己对提示词架构、Skill 架构以及 CLI 工具运行机制的理解，对其进行了全面重构，打造了一个新版 Skill Creator。 在测评环节，我采用 Codex 的 Headless 模式（无需打开图形界面，直接在终端执行）进行了三轮以上的独立对话测试。测试结果显示，新版在各项维度上全面超越原版。
A curated set of personal Claude Code skills, published from ~/.claude/skills/ on my machine. Each top-level folder is a self-contained skill: a SKILL.
Personal agent assistant
是一个专业的 ，用于生成和执行 （Advanced Framework for Simulation, Integration and Modeling）。本项目提供完整的 脚本语言支持，涵盖**、、、、**等场景。
Comprehensive collection of 100+ custom Droids & 300+ Skills for FactoryAI Droid system
VeriFlow-CC: A Claude Code-driven RTL design pipeline. Automates Chip-on-Chat from architecture to synthesis (iVerilog/Yosys) using a stateful, zero-dependency LLM orchestration skill.
A prompt framework for Claude Code that enforces strict TDD, Clean Architecture, and DDD across the full stack.
A coordinated team of 53 Claude Code subagents that build, ship, and operate mobile apps end to end. Each agent owns a narrow slice of the lifecycle so that responsibility, expertise, and review...
OpenCockpit — the open Claude Code GUI for any LLM (Claude, Codex, DeepSeek, GLM, Kimi, Ollama). Cockpit IDE: parallel projects, terminal, browser & DB bubbles, code review.