NOV.mov

🔭 NOVA — Nebula-Oriented Virtual Astronomer

An Agentic RAG Chatbot specialized in Space Exploration & Astronomy.

Built with Flask · OpenAI · ChromaDB · LangChain · Pure HTML/CSS/JS

Quick Start

cd backend
pip install -r requirements.txt
cp .env.example .env   # Add your OPENAI_API_KEY
python main.py
# Open http://localhost:5000

NOVA — Nebula-Oriented Virtual Astronomer

Agentic RAG Chatbot · Space Exploration & Astronomy

---

🗂️ Project Structure

chatbot/
├── backend/
│   ├── data/
│   │   └── space_knowledge.txt     # Knowledge base document
│   ├── chroma_db/                  # Auto-generated vector store
│   ├── main.py                     # Flask server (entry point)
│   ├── rag.py                      # RAG pipeline (embed + retrieve)
│   ├── nova.py                     # NOVA agent logic + 3 scenarios
│   ├── requirements.txt
│   └── .env.example
└── frontend/
    └── index.html                  # Full chatbot UI (HTML/CSS/JS)

---

⚙️ Setup & Run

1. Clone / navigate to the project

cd chatbot/backend

2. Create a virtual environment

python -m venv venv
source venv/bin/activate        # Mac/Linux
venv\Scripts\activate           # Windows

3. Install dependencies

pip install -r requirements.txt

4. Set your OpenAI API key

cp .env.example .env
# Edit .env and add your key:
# OPENAI_API_KEY=sk-...

5. Run the server

python main.py

6. Open in browser

http://localhost:5000

---

🤖 How the Agentic RAG Works

When a user sends a message, NOVA runs a 3-step pipeline:

User Input
    │
    ▼
┌─────────────────────────────────┐
│  STEP 1: Topic Classification   │  ← GPT-4o-mini decides domain
│  "Is this about space/astro?"   │
└────────────┬────────────────────┘
             │
      ┌──────┴──────┐
      │             │
   OUT OF          IN DOMAIN
   DOMAIN             │
      │               ▼
  Scenario 1   ┌─────────────────┐
  (introduce   │ STEP 2: RAG     │  ← ChromaDB similarity search
   NOVA)       │ Retrieve chunks │     (relevance score ≥ 0.3)
               └──────┬──────────┘
                      │
             ┌────────┴────────┐
             │                 │
          FOUND            NOT FOUND
             │                 │
             ▼                 ▼
        Scenario 2        Scenario 3
        (answer from      ("I don't know")
         context)

Scenario Labels (shown in UI badge):

Badge	Meaning
🟡 OUT OF DOMAIN	Topic not related to space/astronomy
🟢 RAG HIT	Found in knowledge base, answered
🔴 RAG MISS	Space topic but not in our data

---

🛠️ Customizing

• Change the topic: Replace space_knowledge.txt in data/ with your own document, update the agent identity in nova.py
• Tune retrieval: Adjust k and the score threshold 0.3 in rag.py
• Rebuild vector store: Delete the chroma_db/ folder and restart the server

---

📦 Tech Stack

Layer	Technology
Frontend	HTML5 + CSS3 + Vanilla JS
Backend	Python
LLM	OpenAI GPT-4o-mini
Embeddings	text-embedding-3-small
Vector Store	ChromaDB (local)
RAG Framework	LangChain