DocuMind — Private Knowledge Base AI

Overview

Your company's knowledge is buried across hundreds of documents — handbooks, policies, product manuals, SOPs. DocuMind turns them into an assistant your team can just ask. Upload your files and DocuMind splits them into passages and indexes them using a local embedding model, so the document text never leaves your infrastructure. When someone asks a question, it finds the most relevant passages and has Claude write a clear answer that cites exactly which documents it used. Because every answer is grounded in your own content, the assistant stays accurate to your policies and won't invent things — if the answer isn't in your documents, it says so. It runs entirely on your own servers with your own Anthropic API key.

Key features

• Ask questions in plain language, get answers grounded in your documents
• Every answer cites the exact source passages it used
• Won't hallucinate — says "I don't know" when the answer isn't in your files
• Local embeddings — your document text never leaves your servers
• Supports PDF, Markdown, and plain text
• Semantic search finds answers by meaning, not just keywords
• Bring your own Anthropic API key — no per-seat limits

Specifications

Product type

Self-hosted software (source code)

License

Perpetual, modify & redistribute in your products

Backend

Python 3.12 · FastAPI

Frontend

Next.js · React

Retrieval

PostgreSQL + pgvector (RAG)

Embeddings

Local model — text stays on your servers

AI model

Claude (Sonnet by default; swappable)

Requirements

Docker + your own Anthropic API key

Delivery

License key + private source download link