Frequently Asked Questions about SearXNG MCP Server

Q: What is an MCP Server? A: MCP stands for Model Context Protocol. An MCP server acts as a bridge, allowing AI models to access and interact with external data sources and tools in a standardized way.

Q: What is SearXNG and why is it used in this MCP Server? A: SearXNG is a privacy-respecting metasearch engine. It’s used because it doesn’t track user queries or IP addresses, ensuring AI agents can search without compromising user privacy.

Q: What are the prerequisites for installing the SearXNG MCP Server? A: You need Python 3.9+, access to a running SearXNG instance, and optionally Docker, uv, or Smithery.

Q: How do I install SearXNG? A: The recommended way is via Docker. A sample docker run command is provided in the documentation.

Q: What is uv and how do I use it for installation? A: uv is a fast Python dependency management tool. You can install it with pip install uv and use it to install the server’s dependencies.

Q: What is Smithery and how do I use it for installation? A: Smithery is a command-line tool for managing AI agent tools and MCP servers. You can install it using pipx install smithery (or another method described in Smithery’s documentation) and then use it to install the SearXNG MCP server.

Q: How do I configure the server? A: Configuration is done via environment variables, which you can set in a .env file. Example .env files are provided.

Q: What are the key environment variables I need to configure? A: The key variables are SEARXNG_BASE_URL (the URL of your SearXNG instance), HOST, PORT, and TRANSPORT (which specifies whether to use SSE or stdio).

Q: What are the different transport protocols (SSE and stdio)? A: SSE (Server-Sent Events) and stdio are different ways the MCP server communicates with AI agents. SSE is typically used for running the server independently, while stdio allows the MCP client to spin up the server.

Q: How do I run the server with SSE transport? A: Set TRANSPORT=sse in your .env file and then run the server script (e.g., uv run dev/searXNG-mcp/server.py).

Q: How do I integrate the server with an MCP client? A: You need to configure your MCP client with the server’s address (for SSE) or the command to run the server (for stdio). Example configurations are provided for different clients and setups.

Q: What are the SearXNG search tool parameters? A: These are parameters you can pass to the search tool to fine-tune your search queries. Examples include q (the query), categories, engines, and language.

Q: Can I build my own MCP server based on this template? A: Yes, this template provides a foundation for building more complex MCP servers. You can add your own tools, lifespan functions, prompts, and resources.

Q: Where can I find more information about SearXNG? A: See the SearXNG documentation for more details.

Q: How does UBOS relate to the SearXNG MCP server? A: UBOS is a full-stack AI Agent Development Platform that provides a broader ecosystem for AI agent development and deployment. The SearXNG MCP Server is available on the UBOS Asset Marketplace.