Provides web search capabilities through a SearXNG instance, allowing for queries across various categories such as news, images, videos, and science with support for language, time range, and safe search filtering.
Click on "Install Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type @ followed by the MCP server name and your instructions, e.g., "@SearXNG MCP Server search for the latest AI news from the past week"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
A Model Context Protocol (MCP) server that provides web search capabilities using SearXNG.
🔍 Web search using your SearXNG instance
🎯 Category filtering (general, news, images, videos, science, etc.)
🌍 Multi-language support
⏰ Time range filtering
🛡️ Safe search options
đź“Š Formatted results with titles, URLs, snippets, and metadata
Related MCP server: SearXNG MCP Server
Node.js 18 or higher
A running SearXNG instance
ngrok (for public sharing)
npm installThe server uses the SearXNG instance at http://192.168.1.44:55001 by default. You can override this with the SEARXNG_URL environment variable:
export SEARXNG_URL=http://your-searxng-instance:portnpm startInstall ngrok if you haven't already:
npm install -g ngrokStart the MCP server:
npm startIn another terminal, expose it via ngrok:
ngrok http 3000Use the ngrok URL in your MCP client configuration
Add this to your LM Studio MCP settings:
{
"mcpServers": {
"searxng": {
"command": "node",
"args": ["C:\\Users\\claym\\Desktop\\public mcp\\index.js"],
"env": {
"SEARXNG_URL": "http://192.168.1.44:55001"
}
}
}
}When sharing with others, they can use this configuration:
{
"mcpServers": {
"searxng": {
"command": "npx",
"args": ["-y", "searxng-mcp-server@latest"]
}
}
}Note: For the npx version to work, you'll need to publish this to npm first.
web_searchSearch the web using SearXNG.
Parameters:
query (required): The search query
categories (optional): Search categories - "general", "news", "images", "videos", "science", etc. Default: "general"
language (optional): Language code like "en", "es", "fr". Default: "en"
time_range (optional): Filter by time - "day", "week", "month", "year". Default: all time
safesearch (optional): Safe search level - 0 (off), 1 (moderate), 2 (strict). Default: 0
Example:
Search for "latest AI news" in the news category from the past week.
├── index.js # Main MCP server implementation
├── package.json # Dependencies and scripts
└── README.md # This fileYou can test the server using any MCP-compatible client like:
LM Studio
Claude Desktop
Other MCP clients
If you can't connect to SearXNG:
Verify your SearXNG instance is running: curl http://192.168.1.44:55001
Check firewall settings
Ensure the URL is correct in the configuration
If searches return no results:
Test the same query directly in SearXNG web interface
Try different categories or remove filters
Check SearXNG logs for errors
MIT
Search the web using SearXNG. Returns up to 10 relevant results with titles, URLs, snippets, and metadata.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | The search query | |
| categories | No | Search categories (e.g., "general", "news", "images", "videos", "science"). Default: "general" | general |
| language | No | Language code (e.g., "en", "es", "fr"). Default: "en" | en |
| time_range | No | Time range filter: "day", "week", "month", "year". Leave empty for all time. | |
| safesearch | No | Safe search level: 0 (off), 1 (moderate), 2 (strict). Default: 0 |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries the full burden. It discloses key behavioral traits: it uses SearXNG, returns up to 10 results with specific fields, and performs searches. However, it misses details like rate limits, authentication needs, error handling, or whether it's read-only/destructive—important for a tool with no annotations.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
The description is front-loaded and efficient: one sentence states the action and engine, another specifies the output format. Every sentence adds critical information without redundancy, making it appropriately sized and well-structured for quick understanding.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given the tool's moderate complexity (5 parameters, no output schema, no annotations), the description is fairly complete: it covers purpose, engine, and output details. However, it lacks behavioral context like rate limits or error handling, and with no output schema, it doesn't fully explain return values beyond a high-level summary.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema description coverage is 100%, providing full parameter documentation. The description adds value by summarizing the tool's purpose and output, but doesn't elaborate on parameter meanings beyond what the schema already covers. With 5 parameters, the baseline is 3 due to high coverage; it earns a 4 for contextualizing the overall search process.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description clearly states the specific action ('Search the web using SearXNG') and resource ('the web'), with precise output details ('Returns up to 10 relevant results with titles, URLs, snippets, and metadata'). It distinguishes itself by specifying the search engine (SearXNG) and result format, which is comprehensive for a standalone tool.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies usage for web searches with result details, but provides no explicit guidance on when to use this tool versus alternatives (e.g., other search methods or tools). With no sibling tools, this is less critical, but it lacks any context on optimal scenarios or limitations.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
With only one tool, there is no possibility of ambiguity or overlap between tools. The tool's purpose is clearly defined and distinct by default.
A single tool inherently has perfect naming consistency, as there are no other tools to compare it against. The name 'web_search' follows a clear verb_noun pattern.
A single tool for a web search server feels thin and limited in scope. While it covers the core search functionality, it lacks additional features like filtering, advanced queries, or result management that might be expected in a more complete search interface.
The server provides only basic search functionality without any supporting operations. There are significant gaps, such as no ability to refine searches, handle pagination, or access search history, making the surface incomplete for robust web search tasks.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Web search, news, page retrieval, sitemaps, and trending topics through Search1API.
Provides AI assistants with access to Seltz's powerful Web Search capabilities.
Search the web, images, videos, news, and local businesses with robust filters, freshness controls…
Search Google straight from your AI agent. Web results, images, videos, news, products, scholarly ar
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/franktheglock/public-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
