HomeProjectsArticlesPK BoardAbout
Explore Projects

MCP Use

Active
GitHub TypeScript MIT

Description

MCP Use is a Model Context Protocol orchestration project that helps agents connect to MCP servers, unify tool invocation, and improve portability across toolchains.

Key Features

  • Fullstack MCP framework with SDK for both TypeScript and Python
  • Build MCP Servers with typed tool definitions and Zod schema validation
  • MCP Apps — interactive React widgets that work across Claude, ChatGPT, and other clients
  • Built-in MCP Inspector for testing and debugging servers
  • One-click deployment to Manufact MCP Cloud with observability and metrics
  • Auto-discovered widget resources — no manual registration needed

Use Cases

💡 Build custom MCP servers exposing domain-specific tools to AI agents
💡 Create interactive MCP Apps (charts, diagrams, maps) for Claude and ChatGPT
💡 Develop and debug MCP toolchains locally before deploying to production
💡 Deploy MCP servers to the cloud with built-in observability and branch previews
💡 Build AI agent tooling with typed schemas and auto-discovered resources

Strengths & Limitations

✅ Strengths

  • • Actively maintained, recent updates
  • • High community interest (10.5k stars)
  • • Permissive open-source license (MIT)
  • • Responsive to issues, low backlog

Tags

mcp tools protocol python

Categories

🔌 MCP Tools

Quick Start

npx create-mcp-use-app@latest → define tool with MCPServer → add widget in resources/ → server auto-discovers widget → deploy to Manufact or self-host
Visit GitHub Visit Website View Docs

Related Projects

Awesome Claude Skills

72.6k · Python
Active A

A curated list of awesome Claude Skills, resources and tools for customizing Claude AI workflows.

mcptoolspython +1
  • · 1000+ production-ready skills - Covering 9 categories including document processing, dev tools, data analysis, marketing, and writing
  • · Cross-agent compatible - Works with Claude Code, Claude.ai, OpenAI Codex, Cursor, Gemini CLI, Windsurf and more
  • · Progressive loading architecture - Each skill loads only name and description (~100 tokens), full content on demand, hundreds coexist
View full details → View on GitHub

Awesome LLM Skills

1.5k · Python
Normal C

A curated list of awesome LLM and AI Agent Skills, resources and tools for customising AI Agent workflows. Works with Claude Code, Codex, Gemini CLI and custom agents.

agenttoolsmcp +2
  • · Curated collection of LLM skills for Claude Code, Codex, Gemini CLI, and more
  • · Covers document processing (docx, pdf, pptx, xlsx), development tools, and research workflows
  • · Cross-platform compatibility with Anthropic, OpenAI, Google, and Alibaba AI tools
View on GitHub

Blender MCP

25.9k · Python
Active A+

MCP server for Blender that enables AI agents to directly control the 3D modeling software for natural language-driven scene creation, model manipulation, and rendering automation.

mcptoolsagent +1
  • · Bidirectional communication — Blender addon creates socket server, MCP Server bridges AI and Blender
  • · 3D object manipulation — create, modify, delete 3D objects in Blender via natural language
  • · Material and rendering control — natural language application of materials, colors, lighting, render settings
View full details → View on GitHub

Related Articles

MCP性能优化stdio

MCP Server Performance: From Protocol to Transport

Are MCP tool calls 10-100x slower than direct HTTP? A systematic deep dive into MCP Server performance bottleneck analysis, protocol-layer optimization (payload size, on-demand tool registration, streaming responses), transport-layer optimization (stdio long connection, HTTP/2, connection pool), tool-internal optimization (async, caching, pre-warming), and deployment-layer optimization.

Read more →