ClawRag

ClawRAG is a self-hosted document question-answering system you run entirely on your own computer or server. It combines document processing with a vector database (a type of database optimized for finding conceptually similar content) and a local AI model, so your documents never need to leave your infrastructure.

You upload files — PDFs, Word documents, text files, CSVs, and more — and the system extracts their content, breaks it into searchable chunks, and stores them in a local vector database called ChromaDB. When you ask a question, it finds the most relevant passages using a hybrid search that combines semantic meaning with keyword matching, then passes those passages to an AI model to generate an answer grounded in your actual documents.

The system runs entirely through Docker (a containerization tool), meaning you can get it running with a single command. It ships with a web interface for browsing and querying, plus a REST API for developers who want to integrate it with other tools.

A free community edition handles PDF, Word, Markdown, text, and CSV files. An enterprise tier (in a separate repository) adds support for more file formats, multiple parallel processing engines that cross-check each other's output for accuracy, graph-based relationship traversal across documents, and multi-tenant isolation.

Intended use cases include legal and compliance document review, medical research analysis, financial due diligence, and technical documentation search. The full README is longer than what was provided.