Provides AI-powered code search via the Exa API with configurable results and JSON metadata.
Configuration
View docs{
"mcpServers": {
"it-beard-exa-server": {
"command": "node",
"args": [
"/path/to/exa-server/build/index.js"
],
"env": {
"EXA_API_KEY": "YOUR_API_KEY"
}
}
}
}You are provided with an MCP server that enables AI-powered code search via the Exa API. It lets you query code repositories and documentation using natural language, returning relevant code examples and metadata in JSON. This server is designed to be connected through an MCP client and configured with your Exa API key to start searching immediately.
You connect to the Exa MCP server through your MCP client and use the search tool to perform natural language queries. When you submit a query, you receive AI-generated results including code examples and relevant documentation. You can adjust the number of results returned to balance breadth and detail.
Prerequisites you need before installation: Node.js and npm installed on your machine.
Clone the server repository and navigate into the project directory.
git clone https://github.com/it-beard/exo-server.git
cd exa-serverInstall project dependencies.
npm installBuild the project to produce the executable runtime.
npm run buildConfigure your Exa API key in the MCP settings to enable authentication for API calls. Create a configuration block for the Exa MCP server as shown.
{
"mcpServers": {
"exa": {
"command": "node",
"args": ["/path/to/exa-server/build/index.js"],
"env": {
"EXA_API_KEY": "your-api-key-here"
}
}
}
}The Exa MCP server exposes a dedicated search tool that you can invoke through your MCP client. Ensure your API key is kept secure and never committed to public version control. You can adjust the number of results returned by the client to tailor the response volume.
Perform an AI-powered search using the Exa API. Accepts a natural language query and a desired number of results, returning AI-generated code examples and documentation.