AskDocs — Minimal RAG Assistant 📄🤖

A tiny Retrieval-Augmented Generation (RAG) assistant: drop your documents into ./data, start the server, and ask questions about them in your browser.

Works with either:

• a local model via Ollama (private, no API key), or
• the OpenAI API.

Features

• Ingest .txt, .md, and .pdf documents
• Chunk + embed documents into a lightweight local index (data/index.json)
• Retrieve the most relevant passages and answer from them only
• Simple chat web UI with source attribution

Quick start

# 1. Install dependencies
pip install -r requirements.txt

# 2. Configure a provider
cp .env.example .env

# 3. Drop your documents into ./data

# 4. Run
python app.py
# open http://localhost:8000

Then click Re-index documents and start asking questions.

Providers

Local with Ollama (default)

ollama pull nomic-embed-text
ollama pull llama3.2

RAG_PROVIDER=ollama

OpenAI

RAG_PROVIDER=openai
OPENAI_API_KEY=sk-...

Project structure

ask-docs/
├── app.py            # FastAPI app + RAG engine
├── static/index.html # chat UI
├── data/             # put your documents here
├── requirements.txt
└── .env.example

License

MIT