Langchain MCP Chat Platform

🚀 Project Introduction

This project is a comprehensive application that integrates multiple services such as natural language processing, web search, and email management. It provides a modular and customizable solution for users to meet different needs.

📦 Installation

Environment Requirements

• Python: It is recommended to use Python 3.8 or higher.
• Node.js: Node.js and npm are required (only for starting Tavily and Gmail services).

Install Dependencies

Install the necessary Python packages in the web_js directory:

pip install -r requirements.txt

⚙️ Configuration

Environment Variables

Create a file named .env and add the following content, replacing it with your actual API keys:

GOOGLE_API_KEY=your_api_key_here
SMITHERY_KEY=your_smithery_cli_key_here
TAVILY_API_KEY=your_tavily_api_key_here  # Optional, uses the SMITHKEY environment variable by default
GMAIL_API_KEY=your_gmail_api_key_here    # Optional, uses the Gmail API key file path by default

🚀 Start the Application

Option 1: Start all MCP services

Run the following command to enable all functions (including muhasebe, Tavily search, and Gmail):

cd web_js
python main.py

Option 2: Start only the muhasebe service (recommended for first-time use)

For initial testing, you can comment out the Tavily and Gmail service configurations in web_js/main.py.

🛠️ Troubleshooting

Application freezes during startup

• Problem: The application becomes unresponsive during startup.
• Solution:
  1. Start only the muhasebe service to test basic functionality.
  2. Ensure that Node.js and npm are installed correctly, and run npm install -g @smithery/cli.

Path issues

• Problem: The MCP server path cannot be found.
• Solution: Use an absolute path to start the MCP service in web_js/main.py, for example:

  "args": ["C:/full/path/to/langchain-mcp-chat-platform/own_mcp/mcp_server.py"]
  

🌐 Third-Party Integration Features

Google Gemini

• Purpose: Natural language understanding and generation.
• Configuration: Ensure that GOOGLE_API_KEY is set correctly.

Tavily

• Purpose: Web search and content extraction.
• Configuration: Specify TAVILY_API_KEY in .env or use the environment variable SMITHKEY.

Gmail

• Purpose: Email operation and management.
• Configuration: Enable the Gmail API in your Google Cloud project and set GMAIL_API_KEY.

⚠️ Notes

1. API key security

   • Add the .env file to .gitignore to avoid exposing sensitive information.
   • Check the .env.example file for the required environment variables.

2. Modular design

   • You can disable specific services as needed. By default, all functions are enabled.

3. Language package version control

   • Use langgraph-prebuilt==0.1.5 to ensure compatibility.
   • Conduct thorough testing before updating dependencies to avoid version compatibility issues.

📄 License

This project is open-source under the MIT license.

📚 Component Details

Import path for create_react_agent

Use the correct import method in your code:

from langgraph.prebuilt.task import create_react_agent

Note: Do not use the following old-style import, as it will cause errors:

from langgraph.prebuilt import create_react_agent