- Home
- MCP servers
- SearXNG
SearXNG
- javascript
0
GitHub Stars
javascript
Language
5 months ago
First Indexed
2 months ago
Catalog Refreshed
Documentation & install
Readme and setup notes from the catalogue, plus a client-ready config you can copy for your MCP host.
You can run a SearXNG-backed MCP Server to perform web searches via MCP clients. It connects to your SearXNG instance, supports category, language, time range, and safe search controls, and returns structured results suitable for automation and LM Studio workflows.
How to use
Use your MCP client to query the searxng MCP Server just like you would with any other MCP endpoint. You provide a search query and optional filters, and your client receives structured results with titles, URLs, snippets, and metadata. For local usage, point your client at the local MCP config and start the server. For public access, you can expose the local MCP instance using a tunneling tool or run an npm-based server wrapper that ships with the MCP setup.
How to install
Prerequisites you need before starting:
# Install dependencies
npm install
Start the MCP server locally after installation. It will host the MCP API that your clients connect to.
npm start
Configuration and usage notes
The MCP server connects to a SearXNG instance. By default it targets http://192.168.1.44:55001, but you can override this with an environment variable.
export SEARXNG_URL=http://your-searxng-instance:port
Additional setup for public access
If you want to share the MCP server publicly while keeping it on your local machine, you can expose it with a tunneling service. Start the MCP server locally and then expose the port using your tunneling tool of choice. The exposed URL becomes the address your MCP clients use.
npm start
# Example using a tunnel tool to expose port 3000
# Run the tunneling service as instructed by the tool
Troubleshooting tips
If you run into connection or no-result issues, verify that the SearXNG instance is reachable from the MCP server host, ensure the SEARXNG_URL is set correctly, and confirm that the MCP server is running. Check the SearXNG logs for any errors that indicate misconfiguration or network problems.
Notes on usage with LM Studio
To use locally in LM Studio, configure the MCP server by specifying the node command and the path to the index.js file. Ensure SEARXNG_URL points to your SearXNG instance.
{
"mcpServers": {
"searxng": {
"command": "node",
"args": ["C:\\Users\\claym\\Desktop\\public mcp\\index.js"],
"env": {
"SEARXNG_URL": "http://192.168.1.44:55001"
}
}
}
}
Public/shared usage via npx (optional)
If you plan to share publicly or with others, you can use a package-based approach to start the MCP server.
{
"mcpServers": {
"searxng": {
"command": "npx",
"args": ["-y", "searxng-mcp-server@latest"]
}
}
}
Available tools
web_search
Search the web using a connected SearXNG instance with configurable categories, language, time range, and safesearch.