# Ihor-Sokoliuk/MCP-SearXNG [Health: Active]

**Category:** 🔎 Search & Data Extraction  
**Repository:** https://github.com/ihor-sokoliuk/mcp-searxng  
**GitHub Stars:** 1378  
**npm Downloads (last month):** 69569  
**Views:** 12  
**Installs:** 0  
**Upvotes:** 0  
**Directory Page:** https://allmcps.com/mcp/ihor-sokoliuk-mcp-searxng

## Description
/☁️ - A Model Context Protocol Server for SearXNG

## Tools
Capabilities this server exposes over MCP:

- **searxng_web_search** — Searches the web using SearXNG and returns a list of results, each with a title, URL, and content snippet. CRITICAL: The required parameter name is exactly `query` (not `prompt`, `q`, or any other name). Calls an external SearXNG instance; availability depends on the `SEARXNG_URL` configuration. Use `pageno` to paginate results; combine `time_range` and `language` to narrow scope. When `engines` and `time_range` are both provided, every selected engine must explicitly advertise time-range support via SearXNG /config; otherwise the request is rejected before search. To read the full text of a result URL, follow up with `web_url_read`.
- **searxng_search_suggestions** — Returns autocomplete suggestions from the configured SearXNG instance. Use this to refine vague or partial queries before searching.
- **searxng_instance_info** — Discovers capabilities from all reachable configured SearXNG instances via /config, including categories.common/available, engines.common/available, defaults, locales, and plugins.
- **web_url_read** — Fetches a URL and returns readable content as markdown. Content-type aware: HTML is converted to markdown; JSON is pretty-printed; plain text, YAML, TOML, and XML are returned as fenced readable text. PDF text extraction is supported with bounded input, output, page count, time, concurrency, and memory; OCR is not supported. Binary, media, archive, and octet-stream downloads other than PDFs are intentionally rejected instead of being returned as raw bytes. When the operator configures browser solvers, mcp-searxng attempts FlareSolverr first and then Byparr only after a busy or transient-unavailable acquisition; cache hits bypass acquisition and a final busy or unavailable provider uses one uncached direct-fetch fallback. Three modes: (1) Full content — omit filtering params; use `startChar`/`maxLength` to paginate large pages. (2) Section extraction — set `section` to return content under a specific heading. (3) Headings only — set `readHeadings: true` to list all headings (mutually exclusive with other filtering params). Returns an error string if the URL is unreachable or content cannot be extracted. Use after `searxng_web_search` to read the full content of individual result URLs.

## Claude Desktop Quick Installation
Install path detected from listing signals. Uses `npx` (confidence: high):

```json
"mcpServers": {
  "mcp-searxng": {
    "command": "npx",
    "args": ["-y","mcp-searxng"],
    "env": {
      "SEARXNG_URL": ""
    }
  }
}
```

**Requires environment variables:** `SEARXNG_URL` — the values above are empty placeholders; fill in real credentials before running (see the repository for what each one is for).

## Documentation

## What Ihor-Sokoliuk/MCP-SearXNG does

Ihor-Sokoliuk/MCP-SearXNG exposes SearXNG search and web-reading functions through the Model Context Protocol. It requires an existing SearXNG instance with JSON search enabled; it connects to that service rather than installing or hosting SearXNG automatically.

The search workflow can begin with query suggestions, continue with a filtered and paginated web search, and then fetch the full content of selected result URLs. The URL reader returns HTML as Markdown and presents JSON in formatted form. Plain text, YAML, TOML, and XML are returned as readable fenced text, while PDF text can be extracted within bounded processing limits.

## How it works

An MCP client starts the server as a local process over STDIO or connects to an independently running Streamable HTTP service. `SEARXNG_URL` identifies the SearXNG backend, not the MCP endpoint. Multiple interchangeable SearXNG replicas can be configured as a semicolon-separated list.

Search requests use the `query` parameter. Pagination is available through `pageno`, while `time_range` and `language` narrow the search. If specific engines and a time range are combined, each selected engine must advertise time-range support in the SearXNG configuration.

URL reading operates directly on the selected website. Large responses can be divided with `startChar` and `maxLength`; a particular heading can be selected with `section`, or headings can be listed with `readHeadings`. These filtering modes are mutually exclusive where applicable.

## Setup and configuration

Install with npm or run through npx. Node.js 22 or later is required for npm and npx installations. Docker images include their Node.js runtime. A minimal client configuration sets `SEARXNG_URL` to the base URL of the chosen SearXNG instance.

```json
{
  "mcpServers": {
    "searxng": {
      "command": "npx",
      "args": ["-y", "mcp-searxng"],
      "env": { "SEARXNG_URL": "https://search.example.com" }
    }
  }
}
```

Leave `MCP_HTTP_PORT` unset for local STDIO operation. HTTP deployments use the MCP service URL and can use static bearer or optional OAuth protection. Docker users must forward the environment variables into the container. Optional browser-solvers can be connected with `FLARESOLVERR_URL`, `BYPARR_URL`, or both.

## Tools and capabilities

- `searxng_web_search` searches SearXNG and returns result titles, URLs, and snippets. It supports pagination, language and time filters, engine selection, and full or compact response formats.
- `searxng_search_suggestions` provides autocomplete suggestions for incomplete or broad queries.
- `searxng_instance_info` reports reachable instance categories, engines, defaults, locales, and plugins from SearXNG configuration data.
- `web_url_read` extracts readable content from known URLs, including HTML, structured text, and bounded PDF text. It can return complete content, a selected section, or headings only.

## Limitations and notes

Search availability depends on the configured SearXNG instance and its enabled engines. A public instance receives the queries sent to it and may log them; using a self-operated instance changes the trust relationship but does not itself provide anonymity.

PDF OCR is not supported. Binary, media, archive, and other octet-stream downloads are rejected instead of being returned as raw bytes. URL reading returns an error string when a target cannot be reached or its content cannot be extracted.

Browser-solvers are optional. When configured, FlareSolverr is attempted before Byparr under the server's failover rules, and a final unavailable or busy provider can trigger one uncached direct-fetch fallback. HTTP session behavior varies by transport mode; legacy stateful sessions should be terminated with `DELETE /mcp` when finished.

_Full upstream README: https://allmcps.com/mcp/ihor-sokoliuk-mcp-searxng/readme_

