AtlasKB

AtlasKB is a full-stack knowledge management platform that lets teams ingest notes, search them in Elasticsearch, and chat with an OpenAI-powered assistant that cites the knowledge base.

The repository contains both the FastAPI backend (auth, KB CRUD, chat/QA workflows, OpenAI + Elasticsearch integration) and the Vite + React frontend (login, chat console, keyword search, KB manager).

---

Feature Overview

Area	Capabilities
Knowledge bases	Create/list/delete, copy UUIDs for binding, export Zip bundles (docs + embeddings) for backup or migration.
Document ingestion	Upload markdown/txt, CSV, DOCX, PPTX, or PDF files to auto-create KB docs (embeddings generated on import).
Document store	Chat/QA turns automatically become Q: / A: documents, chunked and embedded into ES (kb_index, kb_doc_index, kb_doc_embed_index).
Chat workspace	Multi-turn chat with KB binding, rename chats, clear conversation, view referenced snippets, switch between chats.
QA API	retrieves vector-similar chunks and asks OpenAI for an answer, writing results back to the KB.
Keyword search	Dedicated UI + API using ES multi_match with highlighting—no OpenAI call, perfect for deterministic audits.
Auth	JWT login/registration, email-or-username login, bcrypt hashing, precise error handling.

---

Architecture

React + Vite UI  ──────►  FastAPI backend
                         │
                         ├─ OpenAI API (chat + embeddings)
                         └─ Elasticsearch (kb_index, kb_doc_index, kb_doc_embed_index)

• Chat/QA: embed question → fetch top KB chunks via ES dense vectors → append to OpenAI prompt → save answer + embedding.
• Keyword search: call /fulltext-search, which runs ES multi_match + highlighting, returning scored snippets (no OpenAI dependency).
• Export: backend bundles kb.json, docs.json, embeddings.json in Zip; frontend downloads via the Knowledge Bases tab.

---

Getting Started

Prerequisites

• Python 3.10+
• Node.js 18+
• Docker Desktop (for Elasticsearch)
• OpenAI API key

Backend

cd knowledge-base
python -m venv .venv
source .venv/bin/activate        # Windows: .venv\Scripts\activate
pip install -r requirements.txt
cp .env.example .env             # set OPENAI_API_KEY, JWT_SECRET, ES URL

cd deploy
docker-compose up -d             # starts Elasticsearch + Kibana
cd ..

python main.py                   # http://127.0.0.1:8000

Swagger UI: http://127.0.0.1:8000/swagger-ui

Frontend

cd frontend
npm install
cp .env.example .env             # configure VITE_API_BASE if needed
npm run dev                      # http://127.0.0.1:5173

---

Demo

Login & Tabs	Chat Workspace

Keyword Search	Knowledge Bases

---

Enjoy shipping with AtlasKB! Contributions and ideas are always welcome.