title	Mahabharata Character Chatbot
emoji	🏹
colorFrom	yellow
colorTo	indigo
sdk	docker
app_port	7860
pinned	false

Mahabharata Character Chatbot

Talk to the heroes, villains, sages, and queens of the Mahabharata. This app lets you have a conversation with characters like Krishna, Arjuna, Karna, Draupadi, and 23 others — each one speaking in first person, staying in character, and drawing on memories of their own life.

Under the hood, every reply is grounded in a curated set of character "memories" so the answers feel personal and consistent with who that character is, rather than generic.

What it does

• 27 playable characters — including Krishna, Yudhisthira, Bhima, Arjuna, Nakula, Sahadeva, Duryodhana, Karna, Draupadi, Kunti, Gandhari, Bhishma, Drona, Vidura, Shakuni, Ashwatthama, Abhimanyu, Ekalavya, and more.
• Stays in character — each character has its own personality, speaking style, and values. Arjuna is disciplined and warrior-like; Krishna is philosophical and playful; Bhima is loud and passionate.
• Remembers your conversation — your chat with each character is saved, so you can leave and come back and pick up where you left off.
• Web chat interface — a simple browser UI for picking a character and chatting.

Requirements

Before you start you'll need:

1. Python 3.9+ installed on your computer.
2. Two free API keys:
   • A Groq API key (powers the character's responses) — get one at console.groq.com.
   • A Pinecone API key (stores and searches the character memories) — get one at app.pinecone.io.

Both have free tiers that are plenty for personal use.

Setup

Run these steps once from inside the project folder.

1. Install the dependencies

pip install -r requirements.txt

(Optional but recommended — create an isolated environment first:)

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

2. Add your API keys

Create a file named .env in the project folder and paste in your keys:

GROQ_API_KEY=your_groq_key_here
PINECONE_API_KEY=your_pinecone_key_here

3. Load the character memories (one-time)

This uploads the character data so the bots can recall their memories. It also downloads a small AI model the first time, so give it a minute.

python src/ingest.py

You only need to do this once (or again if you edit the memory files in the data folder).

Using the app

Web version (recommended)

python app.py

Then open your browser to http://localhost:5000. Pick a character, type a message, and start chatting. Your conversation is saved automatically, and there's a button to clear history if you want a fresh start.

Command-line version

If you prefer the terminal:

python main.py

Type the name of a character (e.g. Arjuna), then chat. Type quit, exit, or bye to leave.

Tips for a good conversation

• Ask characters about their own lives — their battles, regrets, relationships, and choices. For example, ask Karna about his loyalty to Duryodhana, or Draupadi about the dice game.
• Each character knows their own story best, so questions in first person ("How did you feel when...") tend to give the richest answers.
• Try the same question with different characters to see how their personalities and loyalties color the answer.

Troubleshooting

• "GROQ_API_KEY not found" / "PINECONE_API_KEY not found" — make sure your .env file exists in the project folder and the key names are spelled exactly as above.
• The character doesn't seem to recall anything — make sure you ran python src/ingest.py successfully before chatting.
• Errors mentioning the network or model download — the first run needs an internet connection to download the embedding model and reach the APIs.

Project layout

app.py            Web app (Flask)
main.py           Command-line version
src/              Core logic (character bot, database, data ingestion)
data/             Character memories (*.txt) and personality config
templates/        Web page
static/           Styles and scripts
Books/            Reference source material (not used at runtime)

A note on the source material

The Books folder contains novels and retellings of the Mahabharata used as background reference for shaping the characters. They are not read by the app while it runs — the character knowledge lives in the data folder.