Dir Query uses AI to understand your own documents, so you can ask questions and get answers grounded in what you actually wrote — not just what a model already knows. Each workspace holds one separate collection of documents, for example one per person or company. See it in action:
Yes. Dir Query is open source — you can self-host it on your own documents, adapt it to your own systems, and contribute improvements back.
Test your own workspaceDir Query answers questions with RAG (Retrieval-Augmented Generation), indexed and retrieved with llama-index. The backend is a FastAPI service; the frontend is plain Alpine.js. The project itself is developed using SDD (Spec-Driven Development) — every change starts as a written proposal and spec before any code is written.