MCPHub
MCPHub transforms scattered MCP servers into one unified platform. Connect your AI agent to a single endpoint and instantly unlock unlimited tools—no more juggling dozens of server configs. Hot-swap plugins without restart, proxy existing servers in seconds, and let AI generate new tools from natural language.
Links
README
From the repo.
MCPHub
🚀 An MCP Tool Service Platform - One Connection, Infinite Possibilities for AI Agents
Key Features • Quick Start • Feature Details • API Documentation • Development Guide
📸 System Screenshots
Login Interface

Default credentials: username
admin, passwordadmin123
Module Management Interface

Manage all MCP modules with support for loading, unloading, and reloading operations
Tool Details & Control

Enable/disable each tool individually, edit tool descriptions in real-time
Proxy Server Management

Proxy existing MCP Servers without modification
Usage Statistics Dashboard

Visualize tool call counts, success rates, and response times
AI Code Generation

Generate MCP plugin code through natural language conversation
🎯 What is This?
MCPHub is a tool service platform based on the MCP (Model Context Protocol). It addresses a pain point in current AI Agent systems:
Each MCP Server needs to be configured separately on the client side. As the number of tools increases, configuration management becomes chaotic.
MCPHub's Solution:
┌─────────────────────────────────────────────────────────────────┐
│ Before: Scattered Configuration │
│ Claude/Cursor needs to configure dozens of MCP Server addresses│
│ - mcp-server-filesystem │
│ - mcp-server-github │
│ - mcp-server-postgres │
│ - mcp-server-python │
│ - ... (configuration explosion) │
└─────────────────────────────────────────────────────────────────┘
⬇️
┌─────────────────────────────────────────────────────────────────────┐
│ After: Unified Entry Point
│ Claude/Cursor only needs to configure one address
│ { "mcpServers": { "mcp-hub": { "url": "http://host:6081/mcp" } } }
│
│ MCP Hub automatically aggregates all tools, providing:
│ ✅ Centralized tool management
│ ✅ Hot-pluggable enable/disable
│ ✅ Unified monitoring and statistics
└─────────────────────────────────────────────────────────────────────┘
✨ Key Features
1️⃣ One Address, Infinite Tools
AI Agent clients only need to configure one MCP Server address to access all tools on the platform. Whether built-in tools, custom plugins, or proxied third-party MCP Servers, they are all transparent to clients.
2️⃣ Hot-Pluggable Tool Management
- Module-level Management: Load, unload, and reload modules without restarting the service
- Tool-level Control: Individual tools can be enabled/disabled independently
- Hot Configuration Updates: Tool description changes take effect immediately without restart
3️⃣ Proxy Existing MCP Servers
Already have MCP Servers? No modification needed - just configure a proxy to integrate:
{
"name": "my-existing-server",
"url": "http://existing-server:8080/mcp",
"transport": "streamable-http"
}
The platform automatically discovers and transparently proxies all tools from that server.
4️⃣ AI-Powered MCP Plugin Generation
Built-in AI code generation feature - generate compliant MCP module code through natural language conversation:
User: Help me create a weather query tool
AI: Generated t_weather module with get_current_weather and get_forecast tools...
Everyone can become an MCP tool developer!
5️⃣ Plugin Ecosystem: Install & Share
- One-click Install: Upload a ZIP package to install third-party plugins
- One-click Export: Package custom plugins as ZIP to share with other users
- Modular Isolation: Separate management of built-in tools (
tools/) and external plugins (tools_external/)
6️⃣ Group-Based Access Control
Organize modules and proxy servers into groups to control which tools different AI clients can access:
- Group Isolation: Each group has its own set of modules and proxy servers
- Per-Client Access: Clients specify a group via the
x-group-idHTTP header — they only see and can call tools in that group - Default Group: All newly added modules/proxies are automatically added to the default group, ensuring backward compatibility
- Flexible Assignment: A module or proxy server can belong to multiple groups simultaneously
┌──────────────┐ ┌──────────────┐
│ Group A │ │ Group B │
│ (Dev Tools) │ │ (Ops Tools) │
│ • t_python │ │ • t_cli │
│ • t_notebook│ │ • t_ftp │
└──────────────┘ └──────────────┘
↑ x-group-id: A ↑ x-group-id: B
Claude Dev Agent Claude Ops Agent
7️⃣ Comprehensive Usage Statistics
Intuitively view usage for each tool:
- Call count statistics
- Success rate tracking
- Response time analysis
- 24-hour trend charts
🛠️ Built-in Tools
The platform comes with core tool modules ready to use:
| Category | Module | Tools | Description |
|---|---|---|---|
| 🖥️ CLI Execution | t_cli | run_cli_command, list_directory, get_system_info, get_disk_usage, get_running_processes, find_files... | Execute system commands, view processes, disk monitoring, network info |
| 🐍 Python Execution | t_python | run_python_script, installPackage, run_python_script_async, get_script_status | Safely execute Python scripts, install dependencies, async execution |
| 🤖 GUI Automation | t_autogui | autogui_start_task, autogui_get_status, autogui_get_history, autogui_stop_task | AI-driven desktop automation, automatic mouse and keyboard operations |
| 📓 Notebook | t_notebook | add_note, read_note | Note creation and querying, persistent storage |
Featured: AI-Driven GUI Automation
The t_autogui module implements AI-driven desktop automation:
User: Help me open the browser and visit GitHub
Agent: [Calls autogui_start_task] → AI plans task steps → Executes mouse and keyboard operations
This is a killer feature of this platform: AI can operate the computer desktop like a human, completing complex GUI interaction tasks.
📦 Extension Tools
Beyond built-in tools, the platform supports installing extension plugins. Currently available extension tool in tools_external/:
| Category | Module | Tools | Description |
|---|---|---|---|
| 🌐 FTP Transfer | t_ftp | ftp_upload_file, ftp_download_file, ftp_create_directory | FTP file transfer and management |
Extension tools are located in the
tools_external/directory and can be loaded on demand through the Web interface or API. You can also develop your own extension tools and install them on the platform.
🚀 Quick Start
Requirements
- Python 3.10+
- (Optional) Node.js 18+ - for building the frontend
Installation
# Clone the project
git clone https://github.com/Jayden-Dong/MCPHub.git
cd MCPHub
# Create virtual environment
python -m venv .venv
source .venv/bin/activate # Linux/Mac
# .venv\Scripts\activate # Windows
# Install dependencies
pip install -r requirements.txt
Start the Service
cd app
python main.py
After starting, access:
- 🌐 Web Management Interface:
http://localhost:6080 - 📚 API Documentation:
http://localhost:6080/docs - 🔌 MCP Endpoint:
http://localhost:6081/mcp
Configure MCP Client
Add the following in Claude Desktop / Cursor/Cherry Studio or other MCP clients:
{
"mcpServers": {
"general-tools": {
"url": "http://localhost:6081/mcp"
}
}
}
🎉 Done! Your AI Agent can now use all enabled tools on the platform.
📖 Feature Details
Module Management
Manage modules through the Web interface or API:
┌───────────────────────────────────────────────────────────┐
│ Module List [Scan New Modules] [Install]
├───────────────────────────────────────────────────────────┤
│ ✅ t_cli Loaded [Details] [Reload] [Unload]
│ ✅ t_python Loaded [Details] [Reload] [Unload]
│ ⏸️ t_autogui Paused [Details] [Load]
│ 📦 t_weather Not Loaded [Details] [Load] [Export]
└───────────────────────────────────────────────────────────┘
Tool-level Control
Enter module details to enable/disable each tool individually:
┌─────────────────────────────────────────────────────────┐
│ t_cli Module Details
├─────────────────────────────────────────────────────────┤
│ 🔧 run_cli_command [✅ Enabled]
│ 🔧 list_directory [✅ Enabled]
│ 🔧 get_system_info [❌ Disabled]
│ 🔧 kill_process [❌ Disabled] ⚠️ Dangerous
└─────────────────────────────────────────────────────────┘
Proxy MCP Server
Connect existing MCP Servers to the platform:
Via Web Interface:
- Navigate to the "Proxy Management" page
- Click "Add Server"
- Fill in server name, URL, and transport type
- Click "Sync Tools" → Tools automatically appear in the platform
Via API:
curl -X POST http://localhost:6080/api/proxy/servers \
-H "Content-Type: application/json" \
-d '{
"name": "filesystem-server",
"url": "http://localhost:3000/mcp",
"transport": "streamable-http"
}'
AI Code Generation
Generate MCP plugins through conversation:
# Start conversation
curl -X POST http://localhost:6080/api/codegen/chat \
-H "Content-Type: application/json" \
-d '{"message": "Help me create a stock query tool that supports real-time prices and historical trends"}'
# Preview generated code
curl http://localhost:6080/api/codegen/preview/{task_id}
# Install directly to platform
curl -X POST http://localhost:6080/api/codegen/install/{task_id}
Group Management
Groups allow you to organize modules and proxy servers and control which tools each AI client can access.
Via Web Interface:
- Navigate to the "Group Management" page
- Click "Create Group" and give it a name
- Add modules or proxy servers to the group
- Copy the group ID and pass it to the client via the
x-group-idheader
Via API:
# Create a group
curl -X POST http://localhost:6080/api/groups \
-H "Content-Type: application/json" \
-d '{"name": "dev-tools", "description": "Tools for development"}'
# Add a module to the group
curl -X POST http://localhost:6080/api/groups/{group_id}/modules \
-H "Content-Type: application/json" \
-d '{"module_name": "t_python"}'
Configuring the MCP Client with a Group:
{
"mcpServers": {
"dev-tools": {
"url": "http://localhost:6081/mcp",
"headers": {
"x-group-id": "<your-group-id>"
}
}
}
}
When a client provides an x-group-id header, it can only list and call tools that belong to that group. Clients without a group header can access all enabled tools.
Usage Statistics
View tool call information:
# Overall overview
curl http://localhost:6080/api/stats/overview
# Statistics by tool
curl http://localhost:6080/api/stats/tools
# Recent call records
curl http://localhost:6080/api/stats/recent?limit=100
Response example:
{
"total_calls": 1523,
"success_rate": 0.98,
"top_tools": [
{"name": "run_python_script", "calls": 456, "avg_duration_ms": 1234},
{"name": "run_cli_command", "calls": 312, "avg_duration_ms": 567}
]
}
🔌 API Documentation
Module Management /api/modules
| Method | Path | Description |
|---|---|---|
| GET | /api/modules | Get all module list |
| GET | /api/modules/scan | Scan unloaded modules |
| POST | /api/modules/load | Load module |
| POST | /api/modules/{name}/unload | Unload module |
| POST | /api/modules/{name}/reload | Reload module |
| POST | /api/modules/tool/enable | Enable tool |
| POST | /api/modules/tool/disable | Disable tool |
| POST | /api/modules/install | Install ZIP plugin |
| GET | /api/modules/{name}/export | Export module as ZIP |
| DELETE | /api/modules/{name} | Delete external module |
Proxy Management /api/proxy
| Method | Path | Description |
|---|---|---|
| GET | /api/proxy/servers | Get proxy server list |
| POST | /api/proxy/servers | Add proxy server |
| PUT | /api/proxy/servers/{id} | Update configuration |
| DELETE | /api/proxy/servers/{id} | Delete server |
| POST | /api/proxy/servers/{id}/sync | Sync tool list |
| POST | /api/proxy/servers/{id}/enable | Enable server |
| POST | /api/proxy/servers/{id}/disable | Disable server |
Group Management /api/groups
| Method | Path | Description |
|---|---|---|
| GET | /api/groups | Get all groups |
| POST | /api/groups | Create a group |
| GET | /api/groups/{group_id} | Get group details |
| PUT | /api/groups/{group_id} | Update group info |
| DELETE | /api/groups/{group_id} | Delete group |
| GET | /api/groups/{group_id}/members | Get all members (modules + proxies) |
| POST | /api/groups/{group_id}/modules | Add module to group |
| DELETE | /api/groups/{group_id}/modules/{module_name} | Remove module from group |
| POST | /api/groups/{group_id}/proxies | Add proxy server to group |
| DELETE | /api/groups/{group_id}/proxies/{server_id} | Remove proxy from group |
| GET | /api/groups/by-module/{module_name} | Get groups containing a module |
| GET | /api/groups/by-proxy/{server_id} | Get groups containing a proxy |
Statistics Query /api/stats
| Method | Path | Description |
|---|---|---|
| GET | /api/stats/overview | Overall statistics overview |
| GET | /api/stats/modules | Statistics by module |
| GET | /api/stats/tools | Statistics by tool |
| GET | /api/stats/recent | Recent call records |
AI Code Generation /api/codegen
| Method | Path | Description |
|---|---|---|
| POST | /api/codegen/chat | Conversational code generation |
| POST | /api/codegen/chat/stream | Streaming conversation |
| GET | /api/codegen/preview/{task_id} | Preview generated code |
| POST | /api/codegen/install/{task_id} | Install to platform |
| POST | /api/codegen/download/{task_id} | Download ZIP |
🧩 Development Guide
Creating Custom Modules
For detailed module development specifications, please refer to MCP_MODULE_DEV_GUIDE.md.
Quick Start:
my_module/
├── __init__.py # Package marker
├── MODULE_INFO # Module metadata
├── my_module_mcp.py # MCP tool registration
├── my_module_tool.py # Business logic
└── requirements.txt # Optional dependencies
MODULE_INFO:
{
"display_name": "My Tool",
"version": "1.0.0",
"description": "Tool description",
"author": "Your Name"
}
my_module_mcp.py:
from mcp_service import mcp
from config_py.config import settings
from config_py.prompt import prompt_manager
from utils import com_utils
from utils.com_utils import get_mcp_exposed_url
TOOL_NAME = "my_tool"
@mcp.tool(
name=f'{settings.MCP_TOOL_NAME_PREFIX}{TOOL_NAME}',
description=f'MCP Server URL: {get_mcp_exposed_url()}. {prompt_manager.get(TOOL_NAME)}',
enabled=com_utils.get_tool_enable(TOOL_NAME),
)
def my_tool(param: str) -> dict:
"""Tool description"""
return {"success": True, "data": "result"}
Package & Share
# Package
zip -r my_module.zip my_module/
# Install
curl -X POST http://localhost:6080/api/modules/install \
-F "file=@my_module.zip"
🏗️ System Architecture
┌─────────────────────────────────────────────────────────────────┐
│ MCP Client (Claude / Cursor) │
└───────────────────────────────┬─────────────────────────────────┘
│ MCP Protocol (HTTP Streamable)
▼ :6081
┌─────────────────────────────────────────────────────────────────┐
│ MCP Service (FastMCP)
│ ┌───────────────────────────────────────────────────────────┐
│ │ Middleware Layer
│ │ • RequestMCPMiddleware (logging/statistics)
│ │ • ToolDescriptionCheckerMiddleware (description hot update)
│ │ • ToolEnabledCheckerMiddleware (tool filtering)
│ └───────────────────────────────────────────────────────────┘
│ ┌─────────────┐ ┌─────────────┐ ┌─────────────┐
│ │ Built-in │ │ External │ │ Proxy │
│ │ Tools │ │ Plugins │ │ Tools │
│ │ tools/ │ │ external/ │ │ proxy/ │
│ └─────────────┘ └─────────────┘ └─────────────┘
└─────────────────────────────────────────────────────────────────┘
│
┌─────────────────────────────────────────────────────────────────┐
│ REST API (FastAPI) :6080 │
│ /api/modules /api/stats /api/proxy /api/codegen /api/auth │
└───────────────────────────────┬─────────────────────────────────┘
│
┌─────────────────────────────────────────────────────────────────┐
│ Web Frontend (Vue 3) │
│ Module Management | Tool Control | Proxy Config | Statistics │
└─────────────────────────────────────────────────────────────────┘
│
┌─────────────────────────────────────────────────────────────────┐
│ Data Layer (SQLite) │
│ Module Status | Tool Config | Call Statistics | Proxy Config │
│ User Authentication │
└─────────────────────────────────────────────────────────────────┘
📁 Project Structure
MCPHub/
├── requirements.txt # Python dependencies
├── README.md # This file
├── MCP_MODULE_DEV_GUIDE.md # Module development guide
├── config/
│ ├── config.json # Main configuration file
│ └── mcp_hub.db # SQLite database
└── app/
├── main.py # FastAPI entry point
├── mcp_service.py # MCP service initialization
├── my_mcp_middleware.py # MCP middleware
├── config_py/ # Configuration management
├── managers/ # Core managers
│ ├── module_manager.py # Module lifecycle
│ ├── proxy_manager.py # Proxy servers
│ └── stats_manager.py # Statistics management
├── api/ # REST API
├── tools/ # Built-in modules
│ ├── t_autogui/ # GUI automation
│ ├── t_cli/ # CLI commands
│ ├── t_python/ # Python execution
│ └── ...
├── tools_external/ # External plugins
├── toolsmcp/ # MCP tool wrappers
└── web/ # Vue 3 frontend
⚙️ Configuration
Main configuration file: config/config.json
{
"api": {
"title": "MCPHub Service",
"version": "1.0.0"
},
"server": {
"host": "0.0.0.0",
"port": 6080
},
"mcp": {
"service_name": "mcphub",
"host": "0.0.0.0",
"port": 6081,
"path": "/mcp",
"tool_name_prefix": "",
"module_enable": [
{
"enable": true,
"module_name": "tools.t_cli.cli_mcp",
"disable_tool": [],
"config": {}
}
]
},
"database": {
"db_path": "./config/mcp_hub.db"
},
"auth": {
"secret_key": "your-secret-key",
"token_expire_hours": 24
}
}
| Configuration | Description |
|---|---|
mcp.tool_name_prefix | Tool name prefix for distinguishing multi-instance deployments |
mcp.module_enable | List of modules to auto-load on startup |
module_enable[].disable_tool | List of disabled tool names in the module |
module_enable[].config | Custom configuration passed to the module |
🤝 Contributing
Contributions, bug reports, and suggestions are welcome!
- Fork this repository
- Create a feature branch (
git checkout -b feature/amazing-feature) - Commit your changes (
git commit -m 'Add amazing feature') - Push to the branch (
git push origin feature/amazing-feature) - Create a Pull Request
📄 License
This project is licensed under the MIT License.
🙏 Acknowledgments
- FastMCP - Framework for quickly building MCP Servers
- MCP Protocol - Model Context Protocol specification
- FastAPI - Modern high-performance Python web framework
Empower AI Agents with Infinite Possibilities 🚀
Collected info
- ★ 1 stars
- Language: Python
- Source updated: 3/24/2026
Config for your environment
Replace {MCP_ENDPOINT_URL} with this MCP’s endpoint URL (from its repo or docs above). No API key — you connect directly.
Tool
OS
Config file: ~/.cursor/mcp.json
{
"mcpServers": {
"mcp-server": {
"url": "{MCP_ENDPOINT_URL}"
}
}
}Paste into mcpServers in the config file. Restart Cursor after saving.
If this MCP is also published on mcpchannel.ai, you can subscribe from Browse and use the gateway config there instead.