Mcpez (Proxy Aggregator)

MCPez is a microservice command proxy management platform that unifies the management and standardizes the interfaces of backend services through a web interface, simplifying the integration of AI agents and tools.

Developer tools Artificial intelligence chatbots #Microservice management #AI integration #Standardized interfaces #Docker deployment .Python

rating : 2.5 points

downloads : 13

update time : 2025-04-28

What is MCPez?

MCPez is a microservice command proxy management platform designed to help users easily manage and integrate various backend services, including AI models, local scripts, and remote APIs. It provides a unified web interface to configure and monitor these services and exposes them through standardized proxy interfaces (SSE or STDIO), enabling other applications (especially AI agents) to conveniently call and use them.

How to use MCPez?

Using MCPez is very simple: 1) Quickly deploy the platform via Docker; 2) Create and configure applications in the web interface; 3) Add and manage backend services; 4) Call services through standardized interfaces.

Applicable Scenarios

MCPez is particularly suitable for the following scenarios: AI application development that requires integrating multiple backend services; unified management of scattered scripts and APIs; building multi-agent systems that need to call various tools; rapid testing and deployment of microservice combinations.

Main Features

Web User InterfaceProvides an intuitive graphical interface to manage applications and services, allowing configuration without writing code.

Support for Multiple Service TypesSupports two service types: SSE (proxying remote HTTP SSE services) and STDIO (proxying local command-line processes).

Configuration ManagementSupports importing/exporting JSON configurations and saving common service configurations as templates for easy reuse and sharing.

AI PlaygroundComes with a built-in chat interface, allowing you to configure AI models and use MCPez services as the backend for Tool/Function Calling for easy testing.

Docker SupportProvides a Dockerfile for one-click containerized deployment to ensure environment consistency.

Advantages and Limitations

Advantages

Unified management and standardized interfaces: Simplify service integration and management

Break MCP silos: Easily combine multiple services to create greater value

Local-first and security: Keep sensitive data in the local environment

Dockerized deployment: One-click deployment with consistent environments

Improve service quality: Encourage the integration and sharing of high-quality services

Limitations

Requires a certain technical foundation for configuration and deployment

The current version's features are still being continuously improved

The learning curve may be steep for non-technical users

How to Use

Deploy the Platform

Quickly deploy the MCPez platform using Docker.

Access the Web Interface

Open http://localhost:8088 in your browser to access the management interface.

Create an Application

Create a new application on the editing page and set its name and description.

Add a Service

Add an SSE or STDIO type service to the application and fill in the corresponding configuration.

Save and Start

Save the configuration. For STDIO services, click the start button to run.

Usage Examples

Integrate Multiple AI ServicesConfigure different AI model APIs (such as OpenAI and Gemini) as SSE services and test the calls in the AI Playground.

Automate Local ScriptsConfigure local data processing scripts as STDIO services and call them through standard interfaces.

Frequently Asked Questions

What types of services does MCPez support?

How to ensure the security of service calls?

Can I share the services I configured?

What is the function of the AI Playground?

Related Resources

MIT License

The open-source license used by the project

Docker Documentation

Docker usage guide

SSE Technical Documentation

Server-Sent Events technical reference

🚀 MCPez Project

MCPez is a platform designed to manage various services (both SSE and STDIO types). It offers a web interface that enables users to create, edit, and manage these services, and interact with them through the AI Playground.

✨ Features

Service Management

Main Interface: Displays a list of all created applications and their statuses.
Search: Allows for quick application searches.
Start/Stop: For STDIO-type services, their running status can be directly controlled on the interface.
Edit/View: Users can edit existing services or view detailed information.

Application Configuration

Create New Application: New applications can be created by accessing edit.html.
Import/Export: Supports the JSON file format, facilitating configuration backup and migration for users.
Template Library: Provides a tool template library, enabling users to quickly add common service configurations.

AI Playground

Configure AI Model: Set parameters such as API Key, Base URL, and model name.
Integrate MCP Services: Configure the MCP application address to allow the AI to call relevant services.
Human-AI Interaction: Chat with the AI in the chat interface to experience the tool invocation function.

📦 Installation

Environment Requirements

Operating System: Supports Linux, Windows, and macOS.
Dependencies: Docker and Node.js (version 16+ recommended) need to be installed.

Installation Steps

Clone the project repository:

git clone https://github.com/yourusername/mpez.git
cd mpez

Build the Docker image:

docker build -t mpez .

Start the container:

docker run -d --name mpez-container -p 8088:80 -v mpez_data:/data mpez

💻 Usage Examples

Access the Web Interface

Open http://localhost:8088 (or the host port you specified) in your browser to access the MCPez management interface.

Service Management (`index.html`)

Main Interface: Lists all applications, supporting search and start/stop operations for STDIO-type services.
Edit/View: Enter the detailed configuration page by clicking the corresponding button.

Application Configuration (`edit.html`)

New/Edit: Perform operations by specifying the application ID through URL parameters.
Information Settings: Configure the application name and description.
Server Configuration: Add or modify SSE or STDIO-type services, supporting template import and export.

AI Playground (`chat.html`)

Configure Tools: Set the API information of the AI model and specify the MCP application address.
Interaction Experience: Chat with the AI in the chat interface to test the service invocation function.

📄 License

This project is licensed under the MIT License, which allows users to freely use, modify, and distribute the software while retaining the copyright notice.