Myaiserv

🚀 MCP Server

The MCP Server is a platform that offers a variety of tools and functions, supporting operations such as text analysis, image processing, and weather queries. It can interact via REST API and WebSocket.

🚀 Quick Start

📦 Installation

Install with Poetry

poetry install

After installation, start the project:

poetry run serve

Start with Docker

Use Docker Compose to start the service:

docker compose up -d

✨ Features

Tool Interfaces

• Text Analysis Tools: Support sentiment analysis, keyword extraction, etc.
• Image Processing Tools: Support image scaling, cropping, and filter application.
• Weather Query Tools: Provide weather information based on latitude and longitude.
• Text Processing Tools: Support text formatting, statistical analysis, and entity recognition.

REST API

Get All Tool Lists

curl http://localhost:8000/tools

Execute Specific Tool Operations

curl -X POST http://localhost:8000/tools/{tool_name}

WebSocket Interface

Connect to the WebSocket service:

const socket = new WebSocket("ws://localhost:8000/ws");

socket.onopen = () => {
  socket.send(JSON.stringify({
    type: "initialize",
    id: "my-request-id"
  }));
};

socket.onmessage = (event) => {
  const data = JSON.parse(event.data);
  console.log("Received:", data);
};

GraphQL Interface

Perform query and mutation operations via GraphQL. For example:

# Query all tools
query {
  getTools {
    name
    description
  }
}

# Execute specific tool operations
mutation {
  executeTool(input: {
    name: "text_processor",
    parameters: {
      operation: "statistics",
      text: "Example text for analysis"
    }
  }) {
    content {
      type
      text
    }
    is_error
  }
}

💻 Usage Examples

Testing

Run Tests with Poetry

poetry run pytest

Run Tests with Docker

docker compose up -d && docker compose exec web poetry run pytest

Integrating LLM Models

The MCP Server supports integration with various LLM providers. Here is a usage example:

import httpx

async def query_with_llm(prompt: str):
    async with httpx.AsyncClient() as client:
        # Get information about all tools
        tools_response = await client.get("http://localhost:8000/tools")
        tools = tools_response.json()["tools"]

        # Send a request to the LLM service
        llm_response = await client.post(
            "https://api.example-llm.com/v1/chat",
            json={
                "messages": [
                    {"role": "system", "content": "You have access to the following tools:"},
                    {"role": "user", "content": prompt}
                ],
                "tools": tools,
                "tool_choice": "auto"
            }
        )

        return llm_response.json()

Monitoring and Metrics

The MCP Server provides metrics in Prometheus format, which can be obtained via the following endpoint:

curl http://localhost:8000/metrics

These metrics include, but are not limited to, the number of requests for each tool, request time, and error rate.

Development Guide

Code Formatting and Checking

Use the following commands for code formatting and checking:

just fmt
just lint

📄 License

MIT License