Knowify

m-hannanfaisal/Knowify
★ 1 stars Python AI/LLM Updated today
Agentic RAG chatbot with hybrid retrieval (BM25 + dense), cross-encoder reranking, and a self-correcting retry loop. Long-term memory, MCP client/server support, and RAGAS-evaluated, cited answers — across PDFs, DOCX, Excel, CSV, JSON, HTML, and images.
View on GitHub → 🔍 Audit Wallet Slippage →

Quick Install

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

Add to claude_desktop_config.json:

{
  "mcpServers": {
    "knowify": {
      "command": "uvx",
      "args": [
        "knowify"
      ]
    }
  }
}

Or install with pip: pip install knowify

README Excerpt

<div align="center"> Point Knowify at any product's docs — PDFs, API references, spreadsheets, FAQs — and it answers questions instantly, with the exact source and page cited for every claim. This repo includes a fully working local setup, seeded with the AcmeCRM demo dataset, so you can run it yourself and see exactly how it behaves. Swap the documents and it becomes your assistant, or your client's.

Topics

agentic-aibm25fastapihybrid-searchlangchainlangraphllmmcppythonqdrantragrag-chatbotrag-evaluationragasretrieval-augmented-generation