SearXNG MCP Server

The SearXNG MCP Server bridges the gap between AI assistants and a powerful, privacy‑focused metasearch engine. By exposing SearXNG’s search capabilities through the Model Context Protocol, it lets Claude and other AI agents perform real‑time web queries without leaving their native environment. This eliminates the need for custom API wrappers or manual browser interactions, streamlining workflows that rely on up‑to‑date information.

At its core, the server offers four search modalities: General, Image, News, and Engine Info. A general query aggregates results from all configured engines and categories, giving the assistant a broad view of what the web holds. Image search isolates visual results, enabling agents to fetch relevant pictures for content generation or data analysis. News search introduces a time dimension, allowing users to filter headlines by recency—a crucial feature for trend monitoring or fact‑checking. Finally, Engine Info exposes the catalog of available search engines and categories, letting developers tailor queries to specific domains or privacy settings.

These capabilities are delivered via two transport options— for local command‑line interactions and (Server‑Sent Events) for real‑time streaming over HTTP. The SSE endpoint ( by default) is especially valuable for interactive applications where the assistant needs to present incremental results or progress updates. The server’s configuration flags (, , ) let developers point it at any SearXNG deployment, whether local or hosted, and adjust networking settings to fit their infrastructure.

Typical use cases include:

• Real‑time research assistants that pull current statistics or news articles while drafting reports.
• Content generators that retrieve images or data snippets to enrich articles or social media posts.
• Compliance tools that need to verify facts against multiple search engines without exposing user queries to external APIs.
• Privacy‑centric applications where all traffic remains within an internal network, leveraging SearXNG’s open‑source nature.

By integrating seamlessly with existing MCP workflows—exposing tools that accept a query string and return structured results—the SearXNG MCP Server empowers developers to embed reliable, up‑to‑date search directly into AI pipelines. Its lightweight Go implementation and dual transport modes make it a versatile addition to any developer’s toolchain, offering both simplicity for quick prototypes and robustness for production deployments.