RAG

A friendly AI assistant that understands machine learning, science fiction, and the cosmos. It can answer questions, perform calculations, and search Wikipedia — all directly from your terminal.

---

Quick Start

# Clone and enter the project
git clone https://github.com/bniladridas/rag.git
cd rag

# Setup (virtual environment)
python -m venv venv
source venv/bin/activate  # Windows: venv\Scripts\activate
pip install -e ".[test,dev]"

# Run tests
pytest --cov=src --cov-report=term-missing

Or run with Docker:

docker build -f container/Dockerfile -t rag .
docker run -it rag

With API keys:

docker run -it -e TMDB_API_KEY=your_key -e NASA_API_KEY=your_key rag

---

Capabilities

• Knowledge Areas: Machine learning, sci-fi, and space science
• Built-in Tools: Calculator, Wikipedia search, time/date
• Interface: Simple CLI and TUI modes
• Design: Modular, extensible, and fast (FAISS-powered search)

---

Usage

Interactive Mode

rag

TUI Mode

rag-tui

Examples:

What is deep learning?
Tell me about Interstellar
WIKI: Quantum Computing
CALC: sqrt(144)
TIME:

Data Collection

rag-collect

Fetches data from:

• Machine learning documentation
• TMDB (sci-fi movies)
• NASA (space/astronomy)

---

Configuration

Create a .env file:

TMDB_API_KEY=your_tmdb_api_key
NASA_API_KEY=your_nasa_api_key

Or pass via Docker:

docker run -it -e TMDB_API_KEY=your_key -e NASA_API_KEY=your_key rag

API keys:

• TMDB API
• NASA API

---

Development

# Lint and format
black . && isort . && flake8 && mypy .

# Run tests
pytest

# Local CI / lint scripts
./scripts/run_ci.sh
./scripts/fix_lint.sh

# Build package
python setup.py sdist bdist_wheel

Optional setup:

./scripts/setup_dev.sh

---

Project Structure

src/rag/
├── __main__.py       → CLI entry
├── config.py         → Configuration and API keys
├── data_fetcher.py   → Data collection
├── rag_engine.py     → Core logic
├── tools.py          → Utilities (calc, wiki, etc.)
└── ui/tui.py         → Text-based UI

---

Contribution

1. Fork and branch off main
2. Implement changes
3. Run tests and linters
4. Submit a pull request

Refer to:

• Contributing Guidelines
• CLI Policy
• Changelog

---

Git Hooks & Commit Style

cp scripts/commit-msg .git/hooks/
chmod +x .git/hooks/commit-msg

Commit format:

feat: add feature
fix: resolve issue
docs: update readme

---

Automation

• Dependabot: Keeps dependencies updated
• GitHub Actions: Handles testing and releases
• Release Webpage: release-webpage/index.html

---

License & Security

• Licensed under Apache 2.0 — see LICENSE
• For vulnerabilities, see Security Policy

---

Support

For help or questions, please open an issue on GitHub.