Quirox Telegram Bot Backend MCP Server: Enhancing AI Application Integration

Overview: What is Quirox Telegram Bot Backend MCP Server?

The Quirox Telegram Bot Backend project provides a robust backend infrastructure for integrating AI applications, particularly leveraging Google Gemini (Flash model) to interact with various tools and services. This documentation focuses on the core components and functionalities of the MCP server component, designed for seamless integration with emerging AI platforms.

🔧 Core Features & MCP Capabilities

The MCP Server is a central hub that facilitates communication between an AI model (Google Gemini) and external tools such as Google Sheets, web search engines, and content fetchers. Key capabilities include:

• AI Agent Integration: Utilizes Google Gemini for reasoning and task execution.
• Tool Integration Framework: Offers APIs to connect tools like Google Sheets and DuckDuckGo via MCP protocol.
• MCP Server & Client Interaction: The mcp_server.py runs the MCP server, exposing tools like Sheets and Search. The mcp_client.py handles communication with these services.
• FastAPI Web Interface: Exposes a /chat endpoint for AI interaction through HTTP requests.

⚙️ MCP Architecture & Protocol Implementation

The architecture of the Quirox Telegram Bot Backend centers on the Model Context Protocol (MCP) to standardize interactions between the AI model and the various tools. The protocol ensures that AI applications can efficiently communicate with and utilize external services, enhancing their functionality. Key components include:

• MCP Server (mcp_server.py): Exposes services via MCP framework.
• MCP Client (mcp_client.py): Facilitates communication between the AI model and the MCP server.

MCP Protocol Flow Diagram

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

🚀 Getting Started with Installation

Prerequisites

• Python 3.8+: Required for the installation and execution of Python scripts.
• Google Cloud Platform Services: Access to Google Sheets/Drive API and potentially Gmail API for tool integration.
• API Keys: Gemini API key is necessary for AI model communication.

Steps to Set Up and Run

1. Clone Repository:

   git clone <your-repository-url>
   cd <repository-directory>
   

2. Install Dependencies:

   pip install -r requirements.txt
   

3. Environment Variables Setup:

   Create a .env file in the root directory and add the following variables:

   GEMINI_API_KEY=your_gemini_api_key
   
   # Method 1: Service Account (Recommended for servers)
   SERVICE_ACCOUNT_PATH=path/to/your/service_account.json
   DRIVE_FOLDER_ID=your_google_drive_folder_id
   
   # Method 2: OAuth 2.0 (Requires user interaction for the first run)
   CREDENTIALS_PATH=path/to/your/credentials.json
   TOKEN_PATH=token.json
   
   # Method 3: Base64 Encoded Service Account Config (Alternative)
   CREDENTIALS_CONFIG=your_base64_encoded_service_account_json_string
   

Running the Server Components

1. Run MCP Server:

   python mcp_server.py
   

2. Start FastAPI Application:

   python main.py
   # or using uvicorn directly for more options:
   # uvicorn main:app --reload --host 0.0.0.0 --port 8000
   

3. Connect AI Application to MCP Server: Use the /chat endpoint via POST requests.

💡 Key Use Cases in AI Workflows

Use Case 1: Data Management and Analysis

AI applications can create, update, or fetch data from Google Sheets with ease using the GoogleSheetsService.

Technical Implementation:

• Query: "Create an Excel sheet named 'Test Sheet' and add the headers Name, Age, City in the first row."
• Response: MCP server interacts with the Google Sheets API via ghseet.py, creating a new spreadsheet and adding rows.

Use Case 2: Web Interaction

For web-based tasks such as content fetching or search queries, users can benefit from integrated web services using DuckDuckGo Searcher.

Technical Implementation:

• Query: "Search for 'latest tech news'."
• Response: MCP server uses search_tools.py to fetch and parse the results from DuckDuckGo.

🔌 Integration with MCP Clients

Compatibility Matrix

MCP Client	Resources	Tools	Prompts
Claude Desktop	✅	✅	✅
Continue	✅	✅	✅
Cursor	❌	✅	❌

MCP Configuration Sample

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

📊 Performance & Compatibility Matrix

The Performance and Compatibility Matrix outlines the efficiency and interoperability of the MCP server across different AI clients:

• Claude Desktop: High integration, real-time responses.
• Continue: Seamless tool access with limited chat compatibility.
• Cursor: Primarily data interaction without full AI prompt support.

🛠️ Advanced Configuration & Security

Environment Variables

Update .env file to include specific details for authentication and service configuration. Ensure secure handling of API keys and credentials.

Security Measures

• Secure Authentication: Use OAuth 2.0 or Service Account configurations.
• Rate Limiting: Implement rate limiting mechanisms in search_tools.py for web searches.

FAQs

1. Q: Can other AI applications besides Google Gemini be integrated? A: Yes, the MCP server supports multiple AI applications through custom MCP clients.

2. Q: How do I enable OAuth 2.0 authentication? A: Follow the Google Cloud Platform documentation for detailed steps on setting up OAuth credentials.

3. Q: What is the maximum number of queries per minute? A: Use rate limiting within search_tools.py to control the frequency of requests.

4. Q: How do I contribute to the project? A: Review the contribution guidelines for submitting issues, enhancements, and features.

5. Q: Are there any known limitations with the current version? A: The server currently supports tools without full AI prompt integration, which is being addressed in future updates.

👨‍💻 Development & Contribution Guidelines

To contribute to the project,

1. Fork Repository: Clone or forking allows you to contribute changes.
2. Create Issue/PR: Report bugs, suggest features, and submit pull requests with detailed explanations.
3. Code Style: Follow the existing coding conventions.

🌐 MCP Ecosystem & Resources

Explore additional resources within the Model Context Protocol ecosystem:

• Official Documentation: Detailed guides for setting up and customizing MCP clients.
• Community Forums: Engage in discussions, share experiences, and provide feedback.
• Webinars/Videos: Guided tutorials and case studies from experienced developers.

By positioning this document comprehensively, it becomes a valuable resource for developers aiming to integrate MCP servers within AI workflows.