MyAIServ: AI-Powered FastAPI Server with Model Context Protocol (MCP)

High-performance FastAPI server implementing Model Context Protocol (MCP) for seamless integration with Large Language Models (LLMs). Built with a modern stack, including FastAPI, Elasticsearch, Redis, Prometheus, and Grafana.

Overview: What is MyAIServ MCP Server?

MyAIServ MCP server is an innovative AI-powered FastAPI application that enables developers to integrate various Large Language Models (LLMs) into their applications through the Model Context Protocol. The protocol standardizes interactions between AI applications and external data sources, tools, and prompts, ensuring seamless connectivity and enhanced functionality.

🔧 Core Features & MCP Capabilities

MyAIServ MCP server is built around a comprehensive set of features that make it an indispensable tool for developers working with AI applications. Key features include:

• FastAPI-powered REST, GraphQL, and WebSocket APIs: This integration allows for efficient data exchange and real-time communication between the MyAIServ server and external clients.

• Full MCP Support (Tools, Resources, Prompts, Sampling): By fully supporting these components of MCP, the server ensures a robust connection between AI applications and their required resources.

• Vector Search with Elasticsearch: Utilizing Elasticsearch for vector search capabilities optimizes retrieval speed and accuracy in handling large datasets relevant to AI applications.

• Real-time Monitoring (Prometheus + Grafana): With Prometheus and Grafana integrated into the monitoring stack, developers can track server performance metrics in real time, ensuring optimal application performance.

• Docker-ready Deployment: The server is designed with Docker in mind, making it easy for developers to containerize their applications and deploy them across different environments.

• Comprehensive Test Coverage: Extensive test coverage ensures that the MyAIServ MCP server operates reliably and efficiently under various conditions.

⚙️ MCP Architecture & Protocol Implementation

The implementation of MCP at the core of MyAIServ ensures seamless integration with a wide range of AI applications. The server's architecture is designed to facilitate communication between the client and the backend using the Model Context Protocol.

Mermaid Diagram: MCP Protocol Flow

graph TD
    A[AI Application] -->|MCP Client| B[MCP Protocol]
    B --> C[MCP Server]
    C --> D[Data Source/Tool]
    style A fill:#e1f5fe
    style C fill:#f3e5f5
    style D fill:#e8f5e8

Mermaid Diagram: Data Architecture

graph LR
   S[Server] -->|MCP Message| E[Elasticsearch]
   E --> R[Redis Cache]
   S --> P[Prometheus Metrics]
   P --> G[Grafana Dashboard]
   style S fill:#e5f5dc
   style E fill:#c9daff
   style R fill:#ffe6cc
   style P fill:#b2f5de
   style G fill:#d8e0ee

🚀 Getting Started with Installation

To get started quickly, follow these steps:

1. Clone and Setup:

   git clone https://github.com/eagurin/myaiserv.git
   cd myaiserv
   python -m venv venv
   source venv/bin/activate  # Linux/macOS
   pip install -r requirements.txt
   

2. Configure and Run:

   cp .env.example .env
   uvicorn app.main:app --reload
   

   After running, the API Docs are accessible at http://localhost:8000/docs and GraphQL via http://localhost:8000/graphql.

💡 Key Use Cases in AI Workflows

Use Case 1: Real-time Chatbot Assistance for Customer Support

A customer service team could integrate MyAIServ MCP server into their chatbot application. The real-time monitoring provided by Prometheus and Grafana would enable quick detection of any issues, ensuring the chatbot operates smoothly without interruptions.

# Example Python code snippet for integrating with MyAIServ MCP(server)
from client import MCPClient

client = MCPClient(server='http://localhost:8000')
response = client.request(prompt="How can I assist you today?")
print(response.text)  # Processed response from the LLM

Use Case 2: Content Generation for Marketing Teams

A marketing team could leverage MyAIServ to generate creative content based on specific prompts. By integrating Elasticsearch’s vector search capabilities, they can quickly retrieve and refine suitable data sources, ensuring high-quality content with minimal effort.

# Example Python code snippet for content generation use case
from elasticserach_client import ElasticSearchClient

es = ElasticSearchClient(index='content_sources')
prompts = es.search_vectors(prompt="Marketing campaign slogan")
generated_content = client.request(prompts[0])
print(generated_content.text)  # Suggested marketing slogan

🔌 Integration with MCP Clients

MyAIServ is compatible with a variety of MCP clients, supporting tools and resources required for Large Language Models (LLMs). Below are the current supports:

MCP Client	Resources	Tools	Prompts	Status
Claude Desktop	✅	✅	✅	Full Support
Continue	✅	✅	✅	Full Support
Cursor	❌	✅	❌	Tools Only

📊 Performance & Compatibility Matrix

The server's performance and compatibility are crucial for ensuring optimal usage in diverse AI workflows. The following table summarizes key metrics:

Feature	Performance Metrics	Compatibility (Clients)
API Latency	< 100ms	Claude Desktop, Continue, Cursor
Data Integration	Elasticsearch	Claude Desktop, Continue
Monitoring	Prometheus + Grafana	Full Support

🛠️ Advanced Configuration & Security

MyAIServ offers a wide range of configuration options to customize and secure the MCP server. Environment variables support API key management and other security-related configurations.

Configuration Sample

{
  "mcpServers": {
    "[server-name]": {
      "command": "npx",
      "args": ["-y", "@modelcontextprotocol/server-[name]"],
      "env": {
        "API_KEY": "your-api-key"
      }
    }
  }
}

❓ Frequently Asked Questions (FAQ)

1. Q: How do I configure MyAIServ to work with my specific LLM? A: You can use the provided configuration sample in JSON format to set up your environment variables. Ensure to replace placeholders with actual values.

2. Q: Are there any limitations on the number of MCP clients that can be supported simultaneously? A: MyAIServ is designed to handle multiple MCP clients concurrently, but specific limits might depend on server resources and configurations.

3. Q: What level of support does MyAIServ offer for Cursor, which lacks full Prompt support? A: While full Prompt support from Cursor is not available, the server still supports tools integration, making it a viable option for certain use cases.

4. Q: How can I troubleshoot performance issues in MyAIServ? A: Real-time monitoring via Prometheus and Grafana can help identify performance bottlenecks. Review logs and metrics to pinpoint areas that need optimization.

5. Q: Can I customize the API endpoints for my use case? A: Yes, FastAPI framework allows extensive customization of API endpoints based on specific requirements.

👨‍💻 Development & Contribution Guidelines

Contributions are encouraged to enhance the capabilities of MyAIServ. To contribute:

1. Fork and clone the repository.
2. Develop features or fixes.
3. Ensure tests pass.
4. Submit a pull request.

🌐 MCP Ecosystem & Resources

MyAIServ is part of an expanding ecosystem of MCP-compliant tools and services, fostering interoperability between various AI applications. For more information on the latest updates and community-driven projects, visit our official documentation or join developer forums for ongoing discussions.

By integrating MyAIServ with your AI workflows, you can unlock a host of benefits including enhanced performance, flexibility, and ease of use in working with Large Language Models through MCP.