Reactmcp

🚀 Redefining Search: ReActMCP Web Search (Chinese Version)

ReActMCP Web Search is a powerful web search tool designed specifically for AI assistant frameworks. It offers both basic and advanced web search capabilities and seamlessly integrates with other AI tools through the MCP (Multi-Tool Control Protocol) interface.

Key Features

• Supports basic and advanced web searches.
• An extensible architecture for easy customization and expansion.
• Efficient search algorithms for quick result retrieval.
• Stable and reliable, suitable for various application scenarios.

🚀 Quick Start

ReActMCP Web Search provides a comprehensive set of web search features. Here's a brief overview of how to get started:

Installation

Project Structure

ReActMCP Web Search/
├── servers/                  # Directory for MCP server implementation
│   └── web_search.py        # Main file for web search service
├── requirements.txt         # Required Python package dependencies
└── README.md                # Project documentation

Installation Steps

1. Clone the repository:

git clone https://github.com/yourusername/ReActMCP.git
cd ReActMCP/Web-Search

2. Install dependencies:

pip install -r requirements.txt

Configuration

Environment Variable Configuration

Ensure the following environment variables are set (via the .env file):

EXA_API_KEY=your_api_key_here

Example System Configuration

In web_search.py, you can find the default configuration parameters. Adjust the following as needed:

class WebSearchConfig:
    BASE_URL = "https://api.example.com"
    API_KEY_NAME = "EXA_API_KEY"
    TIMEOUT = 10  # seconds

Usage

Start the Search Server

The MCP server is located in the servers directory. Run the server:

python servers/web_search.py

This command will start the MCP server, listening for requests and providing the following tools:

• search_web: Perform basic web searches.
• advanced_search_web: Conduct advanced web searches with filtering options.

✨ Features

• search_web: Execute basic web searches.
• advanced_search_web: Perform advanced web searches with filtering options.
• Supports multiple search parameters (e.g., keywords, URL limits, etc.).

📦 Installation

Project Structure

ReActMCP Web Search/
├── servers/                  # Directory for MCP server implementation
│   └── web_search.py        # Main file for web search service
├── requirements.txt         # Required Python package dependencies
└── README.md                # Project documentation

Installation Steps

1. Clone the repository:

git clone https://github.com/yourusername/ReActMCP.git
cd ReActMCP/Web-Search

2. Install dependencies:

pip install -r requirements.txt

📚 Documentation

Configuration

Environment Variable Configuration

Ensure the following environment variables are set (via the .env file):

EXA_API_KEY=your_api_key_here

Example System Configuration

In web_search.py, you can find the default configuration parameters. Adjust the following as needed:

class WebSearchConfig:
    BASE_URL = "https://api.example.com"
    API_KEY_NAME = "EXA_API_KEY"
    TIMEOUT = 10  # seconds

Usage

Start the Search Server

The MCP server is located in the servers directory. Run the server:

python servers/web_search.py

This command will start the MCP server, listening for requests and providing the following tools:

• search_web: Perform basic web searches.
• advanced_search_web: Conduct advanced web searches with filtering options.

Testing

In web_search.py, there is a test function test_search() (currently commented out). To run the test, uncomment the test execution block and use Python's asyncio runner:

if __name__ == "__main__":
    import asyncio
    # Uncomment the following line to execute the test search
    # asyncio.run(test_search())
    mcp.run()

This will print the search results for example queries and help you verify that the tool is working correctly.

Claude Desktop Configuration

Add the following to config.json to use this server:

{
  "mcpServers": {
    "websearch": {
         "command": "python",
         "args": ["path/to/servers/exa_web_search.py"]
       }
  }
}

Troubleshooting

• Missing EXA_API_KEY: Ensure the .env file is correctly set with a valid Exa API key.
• Dependency Issues: Check if all required Python packages are installed (see requirements.txt). Reinstall the packages if necessary.
• API Errors: If you encounter errors during web searches, check your network connection and verify the status of the Exa API.

📄 License

This project is released under the MIT License. For more details, see the LICENSE file.

🔧 Contribution

Contributions are welcome! If you have any suggestions, bug fixes, or improvements, please open an issue or submit a pull request.

⭐ Star History