Skip to main content
Glama

MCP Agentic AI Server

by itsDurvank
GitHub
Python
OverviewInspectNewEndpointsSchemaRelated ServersReviewsScore
Need Help?View Source CodeReport Issue

🚀 MCP Agentic AI Server Project

Python Flask Streamlit Gemini License

A comprehensive Model Context Protocol (MCP) implementation featuring dual AI server architecture, real-time monitoring, and an interactive dashboard.

🌟 Project Overview

This project demonstrates a production-ready MCP (Model Context Protocol) Agentic AI Server system with:

  • 🔧 Custom MCP Server - Task-based AI processing with tool integration

  • 🌐 Public MCP Server - Direct AI query processing

  • 🎨 Interactive Dashboard - Real-time monitoring and user interface

  • 📊 Live Statistics - Performance metrics and analytics

  • 🛠️ Extensible Tools - Modular tool framework for custom functionality

🏗️ Architecture

┌─────────────────────────────────────────────────────────────────┐ │ 🎨 Streamlit Dashboard │ │ (Port 8501) │ └─────────────────────┬───────────────────────────────────────────┘ │ ┌─────────────┴───────────────┐ ▼ ▼ ┌───────────────────┐ ┌──────────────────┐ │ 🔧 Custom MCP │ │ 🌐 Public MCP │ │ (Port 8000) │ │ (Port 8001) │ │ │ │ │ │ • Task Creation │ │ • Direct Queries │ │ • Tool Integration│ │ • Simple AI Chat │ │ • Async Processing│ │ • Real-time Stats│ └───────────────────┘ └──────────────────┘ │ │ └─────────────┬─────────────┘ ▼ ┌─────────────────┐ │ 🧠 Google │ │ Gemini API │ └─────────────────┘

🚀 Quick Start

Prerequisites

  • Python 3.12+ (Conda environment recommended)

  • Google Gemini API Key (Get one here)

  • Git for cloning the repository

1. Clone & Setup

# Clone the repository git clone <repository-url> cd mcp_server_project # Create and activate virtual environment (recommended) conda create -n mcp_env python=3.12 conda activate mcp_env # Install dependencies pip install -r requirements.txt

2. Environment Configuration

Create a .env file in the project root:

GEMINI_API_KEY=your_gemini_api_key_here

3. Run the Application

Open 4 terminals and run the following commands:

Terminal 1: Custom MCP Server 🔧

cd mcp-agentic-ai python -m custom_mcp.server

Server will start on

Terminal 2: Public MCP Server 🌐

cd mcp-agentic-ai python -m public_mcp.server_public

Server will start on

Terminal 3: Streamlit Dashboard 🎨

cd mcp-agentic-ai/streamlit_demo streamlit run app.py

Dashboard will open at

Terminal 4: Test the APIs 🧪

# Test Custom MCP Server curl -X POST http://localhost:8000/task \ -H "Content-Type: application/json" \ -d '{"input":"Hello World","tools":["sample_tool"]}' # Test Public MCP Server curl -X POST http://localhost:8001/ask \ -H "Content-Type: application/json" \ -d '{"query":"What is artificial intelligence?"}'

🎯 Features

🔧 Custom MCP Server Features

  • Asynchronous Task Processing - Create tasks with unique IDs

  • Tool Integration Framework - Extensible tool system

  • Performance Monitoring - Real-time statistics tracking

  • Error Handling - Robust error management and logging

🌐 Public MCP Server Features

  • Direct AI Queries - Instant responses from Gemini

  • Simple API - Easy-to-use REST endpoints

  • Statistics Tracking - Performance metrics and analytics

  • High Availability - Designed for concurrent requests

🎨 Dashboard Features

  • Modern UI Design - Glassmorphism effects and animations

  • Real-time Updates - Live statistics and performance metrics

  • Responsive Design - Mobile-friendly interface

  • Interactive Forms - Easy server selection and input handling

📊 API Documentation

Custom MCP Server (Port 8000)

Create Task

POST /task Content-Type: application/json { "input": "Your task description", "tools": ["sample_tool"] } Response: {"task_id": "uuid-string"}

Execute Task

POST /task/{task_id}/run Response: { "task_id": "uuid-string", "output": "AI generated response" }

Get Statistics

GET /stats Response: { "queries_processed": 42, "response_time": 1.23, "success_rate": 95.5, "uptime": 120.5 }

Public MCP Server (Port 8001)

Ask Question

POST /ask Content-Type: application/json { "query": "Your question here" } Response: {"response": "AI generated answer"}

Get Statistics

GET /stats Response: { "queries_processed": 15, "response_time": 0.89, "success_rate": 100.0, "todays_queries": 15 }

🛠️ Project Structure

mcp_server_project/ ├── 📄 README.md # This file ├── 📄 requirements.txt # Python dependencies ├── 📄 .env # Environment variables │ ├── 📁 mcp-agentic-ai/ # Main application │ ├── 📁 custom_mcp/ # Custom MCP server │ │ ├── 📄 server.py # Flask server (Port 8000) │ │ ├── 📄 mcp_controller.py # Business logic │ │ └── 📁 tools/ # Custom tools │ │ └── 📄 sample_tool.py # Example tool │ │ │ ├── 📁 public_mcp/ # Public MCP server │ │ ├── 📄 server_public.py # Flask server (Port 8001) │ │ └── 📄 agent_config.yaml # AI configuration │ │ │ └── 📁 streamlit_demo/ # Interactive dashboard │ └── 📄 app.py # Streamlit app (Port 8501) │ └── 📁 documentation/ # Comprehensive docs ├── 📄 documentation.md # Main documentation ├── 📄 workflows.md # Mermaid workflows ├── 📄 designs.md # Architecture diagrams └── 📄 tech-stack.md # Technology details

🔧 Development

Adding Custom Tools

  1. Create a new tool file in mcp-agentic-ai/custom_mcp/tools/:

# my_custom_tool.py import logging def my_custom_tool(text: str) -> str: """ Your custom tool implementation """ logging.info(f"Processing: {text}") # Your logic here result = text.upper() # Example transformation return result

  1. Import and use in mcp_controller.py:

from custom_mcp.tools.my_custom_tool import my_custom_tool # Add to the run method if "my_custom_tool" in task["tools"]: text = my_custom_tool(text)

Extending the Dashboard

The Streamlit dashboard can be customized by modifying streamlit_demo/app.py:

  • Add new UI components

  • Implement additional statistics

  • Create new visualizations

  • Add export functionality

📚 Documentation

Comprehensive documentation is available in the documentation/ folder:

  • 📄 - Complete project guide (1500+ lines)

  • 📄 - Mermaid workflow diagrams

  • 📄 - System design diagrams

  • 📄 - Technology details

🎓 Learning Outcomes

By completing this project, you'll learn:

  • 🤖 AI Integration - Google Gemini API, prompt engineering

  • 🔧 Backend Development - Flask, REST APIs, microservices

  • 🎨 Frontend Development - Streamlit, modern CSS, responsive design

  • 📊 System Monitoring - Real-time statistics, performance tracking

  • 🏗️ Architecture Design - Microservices, event-driven patterns

  • 🔐 Security Practices - API security, environment management

🚀 Deployment

Local Development

Follow the Quick Start guide above.

Production Deployment

For production deployment, consider:

  • 🐳 Docker - Containerize each service

  • ☸️ Kubernetes - Orchestrate containers

  • 🔒 HTTPS - SSL/TLS certificates

  • 📊 Monitoring - Prometheus, Grafana

  • 🗄️ Database - PostgreSQL, Redis

🤝 Contributing

  1. Fork the repository

  2. Create a feature branch (git checkout -b feature/amazing-feature)

  3. Commit your changes (git commit -m 'Add amazing feature')

  4. Push to the branch (git push origin feature/amazing-feature)

  5. Open a Pull Request

📝 License

This project is licensed under the MIT License - see the LICENSE file for details.

🆘 Support

  • 📚 Documentation - Check the comprehensive docs in /documentation/

  • 🐛 Issues - Report bugs via GitHub Issues

  • 💬 Discussions - Join GitHub Discussions for questions

  • 📧 Contact - Reach out for additional support

🌟 Acknowledgments

  • Google Gemini - For providing excellent AI capabilities

  • Streamlit - For the amazing dashboard framework

  • Flask - For the robust web framework

  • Python Community - For the incredible ecosystem

🎯 Next Steps

  1. 🚀 Run the Application - Follow the Quick Start guide

  2. 📚 Read Documentation - Explore the comprehensive docs

  3. 🔧 Customize Tools - Add your own custom tools

  4. 🎨 Enhance UI - Improve the dashboard design

  5. 📊 Add Features - Implement new functionality

  6. 🚀 Deploy - Take it to production

Ready to build the future of AI? Let's get started! 🚀

Built with ❤️ for the AI community. Star ⭐ this repo if you find it helpful!

This server cannot be installed

-
security - not tested
F
license - not found
-
quality - not tested

How are these scores calculated?

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

A dual-server MCP implementation with task-based AI processing using Google Gemini API, featuring tool integration, real-time monitoring dashboard, and extensible framework for custom AI workflows.

  1. 🌟 Project Overview
    1. 🏗️ Architecture
      1. 🚀 Quick Start
        1. Prerequisites
        2. 1. Clone & Setup
        3. 2. Environment Configuration
        4. 3. Run the Application
      2. 🎯 Features
        1. 🔧 Custom MCP Server Features
        2. 🌐 Public MCP Server Features
        3. 🎨 Dashboard Features
      3. 📊 API Documentation
        1. Custom MCP Server (Port 8000)
        2. Public MCP Server (Port 8001)
      4. 🛠️ Project Structure
        1. 🔧 Development
          1. Adding Custom Tools
          2. Extending the Dashboard
        2. 📚 Documentation
          1. 🎓 Learning Outcomes
            1. 🚀 Deployment
              1. Local Development
              2. Production Deployment
            2. 🤝 Contributing
              1. 📝 License
                1. 🆘 Support
                  1. 🌟 Acknowledgments
                    1. 🎯 Next Steps

                      New MCP Servers

                      MCP directory API

                      We provide all the information about MCP servers via our MCP API.

                      curl -X GET 'https://glama.ai/api/mcp/v1/servers/itsDurvank/Mcp_server'

                      If you have feedback or need assistance with the MCP directory API, please join our Discord server