ReActMCP
ReActMCP Web Search is a Model Context Protocol server designed to integrate robust web search features into AI assistants. It uses the Exa API for performing searches and delivers results in a user-friendly format. The project is part of a broader ReActMCP initiative aiming to provide diverse AI capabilities.
What is the primary function of ReActMCP Web Search?
ReActMCP Web Search integrates web search capabilities into AI assistant frameworks, allowing for real-time, markdown-formatted search results using the Exa API.
What are the requirements to run ReActMCP Web Search?
You need Python 3.8+, python-dotenv, and the exa_py client, along with other dependencies required by your MCP framework.
How do I configure the environment for ReActMCP Web Search?
Create a .env file with your EXA_API_KEY and other necessary API keys, and configure the MCP settings in the mcp_config.json file.
How can I test the web search capabilities?
You can test the search capabilities by running the test function in web_search.py using Python's asyncio runner.
What should I do if I encounter API errors?
Check your network connection, verify the Exa API status, and ensure that your .env file is correctly set up with a valid Exa API key.