Allows searching the web via a self-hosted SearXNG instance, returning structured results with title, URL, and snippet.
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 search the web for latest climate change research"
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.
MCP server for web search via a self-hosted SearXNG instance.
Provides a single search_web tool that queries your SearXNG instance and returns structured results (title, URL, snippet).
export SEARXNG_BASE_URL="http://searxng:8080"Optional:
export SEARXNG_LANGUAGE="en-US"
export SEARXNG_CATEGORIES="general"
export SEARXNG_ENGINES="google,bing"
export SEARXNG_TIME_RANGE="month"
export SEARXNG_SAFESEARCH="1"
export SEARXNG_USER_AGENT="my-app/1.0"
export SEARXNG_HEADERS_JSON='{"Authorization":"Bearer ..."}'
export SEARXNG_TIMEOUT_SECONDS="30"No install needed — run directly from GitHub with uvx:
uvx --from "git+https://github.com/fighter3005/search-mcp" searxng-mcpOr from a local clone:
uvx --from /path/to/search-mcp searxng-mcpRelated MCP server: Basic MCP Tools
claude mcp add --transport stdio searxng-search \
-e SEARXNG_BASE_URL="http://searxng:8080" \
-- uvx --from "git+https://github.com/fighter3005/search-mcp" searxng-mcpcodex mcp add searxng-search \
--env SEARXNG_BASE_URL="http://searxng:8080" \
-- uvx --from "git+https://github.com/fighter3005/search-mcp" searxng-mcpCreate .cursor/mcp.json or .vscode/mcp.json:
With uvx from GitHub:
{
"mcpServers": {
"searxng-search": {
"command": "uvx",
"args": ["--from", "git+https://github.com/fighter3005/search-mcp", "searxng-mcp"],
"env": {
"SEARXNG_BASE_URL": "http://searxng:8080"
}
}
}
}After install:
{
"mcpServers": {
"searxng-search": {
"command": "searxng-mcp",
"env": {
"SEARXNG_BASE_URL": "http://searxng:8080"
}
}
}
}Edit claude_desktop_config.json:
With uvx from GitHub (no install needed):
{
"mcpServers": {
"searxng-search": {
"command": "uvx",
"args": ["--from", "git+https://github.com/fighter3005/search-mcp", "searxng-mcp"],
"env": {
"SEARXNG_BASE_URL": "http://searxng:8080"
}
}
}
}After install (pip install -e .):
{
"mcpServers": {
"searxng-search": {
"command": "searxng-mcp",
"env": {
"SEARXNG_BASE_URL": "http://searxng:8080"
}
}
}
}Add to your config.yaml:
mcp_servers:
searxng-search:
command: uvx
args:
- "--from"
- "git+https://github.com/fighter3005/search-mcp"
- "searxng-mcp"
env:
SEARXNG_BASE_URL: "http://searxng:8080"Or as a JSON litellm_config:
{
"mcp_servers": {
"searxng-search": {
"command": "uvx",
"args": ["--from", "git+https://github.com/fighter3005/search-mcp", "searxng-mcp"],
"env": {
"SEARXNG_BASE_URL": "http://searxng:8080"
}
}
}
}Tool | Parameters | Description |
|
| Search the web via SearXNG |
Returns {"results": [{"title": str, "url": str, "snippet": str}, ...]}.
Python 3.11+
A running SearXNG instance with JSON format enabled
Search the web using SearXNG.
Args: query: Search query string. num_results: Number of results to return (default 5).
Returns: {"results": [{"title": str, "url": str, "snippet": str}, ...]}
Raises: If SEARXNG_BASE_URL is not configured or the instance returns an error.
| Name | Required | Description | Default |
|---|---|---|---|
| query | Yes | ||
| num_results | No |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
No annotations are provided, so the description carries full burden. It discloses the return format and error conditions (configuration or instance errors) but omits other behavioral traits such as read-only nature, rate limits, or timeouts.
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 concise, well-structured, and front-loaded. It uses a clear Args/Returns/Raises format without redundant information, making efficient use of space.
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?
For a simple search tool with no output schema, the description covers all necessary aspects: purpose, parameters, return format, and error conditions. It is contextually complete for an AI agent.
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 0%, and the description fully compensates by documenting both parameters: 'query' as 'Search query string' and 'num_results' as 'Number of results to return (default 5)'. This adds complete meaning beyond the schema.
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 tool's purpose: 'Search the web using SearXNG'. The verb 'Search' and resource 'web' are specific and unambiguous. With no sibling tools, no differentiation is needed.
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 implicitly indicates when to use this tool (when a web search is needed). Although no sibling tools exist, it could be improved by explicitly mentioning appropriate use cases or prerequisites like SearXNG configuration.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Only one tool exists, so there is no possibility of confusion between tools.
The single tool follows a clear verb_noun pattern (search_web), and there are no other tools to create inconsistency.
One tool is on the low end for a typical server, but it is focused on a single search operation, making it borderline appropriate.
The server covers the basic search use case, but lacks additional search types or configuration options that might be expected from a full SearXNG integration.
Unclaimed servers have limited discoverability.
If you are the server author, to access and configure the admin panel.
Docs: https://docs.keenable.ai/mcp-server Keenable is a free, remote MCP server that gives agents access to the web index. Search the web with ranked results and date/site filters, then fetch any indexed page as clean markdown. Works out of the box with no account or API key.
Serper MCP — wraps the Serper Google Search API (serper.dev)
MCP server for Google search results via SERP API
We provide all the information about MCP servers via our MCP API.
curl -X GET 'https://glama.ai/api/mcp/v1/servers/fighter3005/search-mcp'
If you have feedback or need assistance with the MCP directory API, please join our Discord server
