What is Mcp Use Demo

mcp-use-demo is a demonstration project that showcases the implementation of a Large Language Model (LLM) Agent with tool access using MCP-Use and Groq.

Use cases

Use cases for mcp-use-demo include developing chatbots, enhancing customer support systems, and creating interactive educational tools that utilize LLM capabilities.

How to use

To use mcp-use-demo, clone the repository, set up a virtual environment, install the required dependencies, create necessary environment files with your API keys, and run the application using the command ‘bashuv run app.py’.

Key features

Key features include integration with Groq’s LLMs, tool access via MCP (Model Context Protocol), web search capabilities using the Brave Search API, conversation memory for context-aware responses, and an interactive chat interface.

Where to use

mcp-use-demo can be used in fields such as artificial intelligence, customer service automation, and interactive applications that require natural language understanding and generation.

MCP-Use LLM Agent Demo

A demonstration of implementing an LLM Agent with tool access using MCP-Use and Groq.

Features

• Integration with Groq’s LLMs
• Tool access via MCP (Model Context Protocol)
• Web search capabilities using Brave Search API
• Conversation memory for context-aware responses
• Interactive chat interface

Prerequisites

• Python 3.8+
• Node.js (for running certain MCP servers)
• Groq API key
• Brave Search API key

Installation

1. Clone this repository

   git clone https://github.com/vineetvk19/mcp-use-demo.git
   cd mcp-use-demo
   
   

2. Set up a virtual environment

   bashuv venv
   source .venv/bin/activate  # On Windows: .venv\Scripts\activate
   
   

3. Install dependencies

   bashuv add mcp-use langchain-groq python-dotenv
   
   

Create environment files:

• Create a .env file with your Groq API key
• Create a mcp.json file with your Brave Search API configuration

  {
  "mcpServers": {
    "brave-search": {
      "command": "npx",
      "args": [
        "-y",
        "@modelcontextprotocol/server-brave-search"
      ],
      "env": {
        "BRAVE_API_KEY": "<BRAVE-API-KEY>"
      }
    }
  }
  
  
  

Usage

Run the application:

bashuv run app.py