@semidark/mcp-litellm-searxng
Provides web search capabilities through a LiteLLM proxy backed by SearXNG, enabling AI agents to search the web and retrieve results.
Click on "Deploy 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., "@@semidark/mcp-litellm-searxng search for latest developments in renewable energy"
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.
@semidark/mcp-litellm-searxng
MCP server for searching the web through a LiteLLM proxy's SearXNG-backed search endpoint.
Installation
No local installation required — the server is executed on-demand via npx:
{
"mcp": {
"azurelit-search": {
"type": "local",
"command": ["npx", "-y", "@semidark/mcp-litellm-searxng"],
"environment": {
"LITELLM_SEARCH_URL": "https://your-litellm-proxy/v1/search/searxng-search",
"LITELLM_API_KEY": "sk-your-key",
"LITELLM_TIMEOUT_MS": "30000"
}
}
}
}With OpenCode, you can usually reuse your existing secret file:
{
"mcp": {
"azurelit-search": {
"type": "local",
"command": ["npx", "-y", "@semidark/mcp-litellm-searxng"],
"environment": {
"LITELLM_SEARCH_URL": "https://litellm-proxy.example.com/v1/search/searxng-search",
"LITELLM_API_KEY": "{file:~/.secrets/azure-lit.key}",
"LITELLM_TIMEOUT_MS": "30000"
}
}
}
}Related MCP server: mcp-searxng
Development
Run the server locally:
npm startRun the integration test harness:
LITELLM_SEARCH_URL="https://your-litellm-proxy/v1/search/searxng-search" \
LITELLM_API_KEY="sk-your-key" \
npm testnpm test is an integration test against a real LiteLLM endpoint. It does not mock the network and fails fast if LITELLM_SEARCH_URL or LITELLM_API_KEY is missing.
Environment Variables
Variable | Required | Default | Description |
| Yes | — | Full URL to the LiteLLM search endpoint |
| Yes | — | Bearer token for API authentication |
| No |
| Request timeout in milliseconds |
Tools
web_search
Search the web using the configured LiteLLM proxy search backend.
Parameters:
query(string, required): Search querymax_results(number, optional, default 10): Max results to return (1–50)
Returns:
text output formatted for chat clients
structuredContent.resultsas the raw LiteLLM result array
License
MIT
Available Tools
1 toolweb_searchWeb SearchA
Search the web using the LiteLLM proxy's configured search backend. Returns ranked search results with titles, URLs, snippets, and dates.
Inputs: query (required), max_results.
Tool Schema Changelog
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
Loading tool schema changelog…
TDQS
Scored across 1 tool
With only one tool, there is no possibility of confusion between tools. The single tool has a clear, distinct purpose.
The sole tool 'web_search' follows a clear verb_noun pattern. Consistency is not an issue with a single tool.
A single tool is minimal but acceptable for a focused search server. However, it feels thin compared to typical MCP servers.
The server provides exactly what its name implies: web search. For a narrow-purpose tool, it covers the core functionality without obvious gaps.
Related MCP Connectors
Provides AI assistants with access to Seltz's powerful Web Search capabilities.
LLM-ready web search + instant answers + URL-to-clean-text fetch for agents and RAG.
Enable AI assistants to perform web searches using Perplexity's Sonar Pro.
The best web search for your AI Agent
Related MCP Servers
- AlicenseAqualityDmaintenanceEnables AI assistants to perform web searches and read URL content via a SearXNG instance.213 npmMIT
- AlicenseNot gradedqualityDmaintenanceIntegrates SearXNG API to give AI assistants web search and URL reading capabilities.13 npmMIT
- AlicenseAqualityDmaintenanceEnables local LLMs to search the web and fetch clean content from URLs without API keys, using SearxNG and Mozilla Readability.242MIT
- FlicenseNot gradedqualityDmaintenanceEnables web search and content scraping from multiple engines via a local SearXNG instance, allowing AI assistants to retrieve and extract web content.1-