mcp-local-rag
by nkapila6·★ 125·Score 47
Local RAG web search MCP server with 9+ search backends and no API keys required.
Overview
mcp-local-rag is a privacy-focused, local-first RAG-like web search MCP server that runs entirely without external API dependencies. It supports multiple search engines including DuckDuckGo, Google, Bing, Brave, Wikipedia, and others, with semantic similarity ranking to deliver relevant results. The server is containerized and works with various MCP clients like Claude Desktop, Cursor, and Goose. It includes Agent Skills to teach LLMs best practices for effective web searches and comprehensive deep research capabilities.
Try asking AI
After installing, here are 3 things you can ask your AI assistant:
When to choose this
Choose this when you need privacy-focused local web search without API dependencies and require multiple search backends for comprehensive research.
When NOT to choose this
Not recommended for production environments requiring high reliability or if you need web access that requires authentication.
Tools this server exposes
5 tools extracted from the READMEdeep_researchComprehensive multi-engine research across multiple search backends
deep_research_googleGoogle-focused deep dive with technical/scientific queries
deep_research_ddgsPrivacy-first deep research using DuckDuckGo
rag_search_ddgsQuick single search using DuckDuckGo for fast answers
rag_search_googleQuick single search using Google for focused results
Comparable tools
Installation
Installation
**Using Docker (recommended)**
{
"mcpServers": {
"mcp-local-rag": {
"command": "docker",
"args": [
"run",
"--rm",
"-i",
"--init",
"-e",
"DOCKER_CONTAINER=true",
"ghcr.io/nkapila6/mcp-local-rag:v1.0.2"
]
}
}
}**Using uvx**
{
"mcpServers": {
"mcp-local-rag":{
"command": "uvx",
"args": [
"--python=3.10",
"--from",
"git+https://github.com/nkapila6/mcp-local-rag",
"mcp-local-rag"
]
}
}
}Compare mcp-local-rag with
Last updated · Auto-generated from public README + GitHub signals.