kapa-inspired-rag-mcp

AyanArshad02/kapa-inspired-rag-mcp

★ 0 stars Jupyter Notebook AI/LLM Updated 1mo ago

Production-grade documentation RAG system with multi-source ingestion, hybrid retrieval, freshness handling, tenant isolation, and MCP server, built with the depth and tradeoffs

View on GitHub → Try with Claude — $10 free →

Quick Install

Copy the config for your editor. Some servers may need additional setup — check the README.

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "kapa-inspired-rag-mc": {
      "command": "npx",
      "args": [
        "-y",
        "AyanArshad02/kapa-inspired-rag-mcp"
      ]
    }
  }
}

README Excerpt

> A production-grade, multi-tenant documentation RAG system with empirically validated retrieval, streaming answers, and an MCP server — built from first principles to match what real AI infrastructure companies ship. --- Most RAG tutorials stop at "chunk → embed → retrieve → generate." This project goes further — it's what a real company like [kapa.ai](https://kapa.ai) actually has to build:

Tools (7)

CodeBlockAwareChunkerDELETEGETHeadingAwareChunkerHierarchicalChunkerPOSTThreadAwareChunker

kapa-inspired-rag-mcp

Quick Install

README Excerpt

Tools (7)

Related AI/LLM Servers