MCP based AI Assistant

This is an MCP based AI Assistant built using Groq, LangChain, and MCP, with a Streamlit frontend and FastAPI backend. It enables chat interactions powered by large language models, enhanced with optional tools like search agents and browser automation.

---

Features

• Groq LLM with LangChain and memory enabled conversation
• MCP server such as Google Search and Playwright integration
• FastAPI backend to handle chat API requests
• Streamlit frontend for real-time chat
• Utilized Docker for containerized deployment

---

Project Structure

MCP_Assistant/
├── app.py 
├── main.py # FastAPI server with chat endpoint
├── streamlit_app.py # streamlit frontend UI
├── browser_mcp.json # MCP servers
├── .env # Environment variables (API keys)
├── requirements.txt # required dependencies
└── README.md # You are here!

---

Setup Instructions

1. Clone and Install Dependencies

git clone https://github.com/itsabhishekm/MCP_Assistant.git
cd MCP_Assistant
python -m venv .venv
.venv\Scripts\activate   
pip install -r requirements.txt

2. Set Environment Variables

In the .env file in the root past your groq API KEY: GROQ_API_KEY=your_groq_api_key

3. Configure MCP Tools (Not Required but optional)

If you want to add any other MCP server customize browser_mcp.json:

  "mcpServers": {
    "google-search": {
      "command": "npx",
      "args": ["-y", "@mcp-server/google-search-mcp@latest"]
    },
    "playwright": {
      "command": "npx",
      "args": ["@playwright/mcp@latest"]
    }
  }
}

Or if you don't want any MCP server, just leave it empty:

{ "mcpServers": {} }

Running the app

1. Start FastAPI Backend

uvicorn main:app --reload --port 8000

2. Launch Streamlit Frontend

streamlit run streamlit_app.py Visit: http://localhost:8501