Best for
- Users who want self-hosted search infrastructure
- Teams with an existing SearXNG instance and privacy requirements
- MCP clients that need paginated search and URL reading
S
61FMRS · C
Based on the repository description, server manifest, and README, this is a third-party MIT-licensed SearXNG MCP server at version 1.12.0. It uses stdio by default and also supports optional Streamable HTTP transport.
SearXNG Search is a standalone MCP server that connects one or more SearXNG instances to AI assistants. It provides web search, query suggestions, instance capability discovery, and URL content reading, with pagination, filters, caching, failover, and optional HTTP transport.
Tools
searxng_web_search
Runs web searches with pagination, time-range, language, safe-search, relevance, category, and engine filters, returning formatted text or raw JSON.
searxng_search_suggestions
Gets autocomplete suggestions from SearXNG for a partial or complete query.
searxng_instance_info
Discovers categories, engines, defaults, locales, and plugins exposed by reachable SearXNG instances.
web_url_read
Fetches URL content and converts it to Markdown; supports HTML, JSON, plain text, section and paragraph extraction, and heading-only output.
Setup
Add the install configuration to your MCP client and replace SEARXNG_URL with the URL of your SearXNG instance. Global npm installation, Docker, and Docker Compose are also documented. SEARXNG_URL is the only required variable. For HTTP mode, set MCP_HTTP_PORT and connect to the /mcp endpoint.
claude_desktop_config.json
{"mcpServers":{"searxng":{"command":"npx","args":["-y","mcp-searxng"],"env":{"SEARXNG_URL":"YOUR_SEARXNG_INSTANCE_URL"}}}}
Fit and risk
Not for
- Users without an accessible SearXNG instance
- Users seeking a native SearXNG plugin installation
- Workflows that require reading PDFs, media, archives, or other binary downloads
Required permissions
- Requires access to the configured SEARXNG_URL, which the manifest marks as required and secret.
- May use AUTH_USERNAME and AUTH_PASSWORD for legacy global SearXNG Basic Auth.
- Search queries are passed to external search services, and web_url_read makes requests to user-specified URLs.
Risks and side effects
- Privacy depends on the connected SearXNG instance and its configuration.
- Search queries may be passed to external services used by SearXNG.
- URL reading creates outbound requests; private or internal URLs and redirects are blocked by default, but network deployments should still enable and configure hardening.
- HTML fallback depends on the instance's page structure and may return fewer results without relevance scores or engine names.
Troubleshooting
- Confirm that SEARXNG_URL points to a reachable SearXNG instance.
- For 403 errors, verify that SearXNG search.formats includes json; if the instance cannot be changed, enable SEARXNG_HTML_FALLBACK=true.
- Check whether the instance requires AUTH_USERNAME and AUTH_PASSWORD.
- For HTTP deployments, check MCP_HTTP_PORT and MCP_HTTP_HOST, then verify the service through /health.
- For TLS failures, check corporate proxy or custom CA configuration.
Use cases
Perform web, news, and article searches through SearXNG from an AI assistant
Read web pages and extract sections, paragraphs, or headings
Use interchangeable SearXNG replicas for failover or parallel fan-out
Inspect the categories and engines supported by configured instances
Supported clients
Claude DesktopFull support
CursorFull support