Crypto AI Agent

A conversational AI chatbot built with Python and Flask that can answer general questions, fetch real-time cryptocurrency data (prices, market cap, trading volume, OHLC, top gainers/losers), and even predict future prices using simple machine learning. The agent uses NLTK for natural language processing, CoinGecko API for crypto data, and scikit-learn for basic regression.

Features

Conversational Chatbot: Handles greetings, small talk, and fallback responses using NLTK's pattern-matching.
Crypto Price Lookup: Get the latest USD price for any supported cryptocurrency.
Market Cap & Volume: Query the current market capitalization and trading volume.
Top Gainers & Losers: See the top performing and worst performing coins (requires CoinGecko Pro API).
OHLC Data: Retrieve Open, High, Low, Close data for a coin.
Coin Recommendation: Suggests a coin with the highest market cap and volume.
Price Prediction: Uses linear regression on the last 30 days to predict the next day's price.
Web Interface: Simple chat UI built with Flask, HTML, and CSS.

Project Structure

.
├── app.py                # Main Flask backend and AI logic
├── requirements.txt      # Python dependencies
├── .env                  # Environment variables (API keys)
├── templates/
│   └── index.html        # Chatbot web interface
├── static/
│   └── css/
│       └── style.css     # UI styling
├── reqs.py               # (Optional) Dependency helper
└── myenv/                # Python virtual environment (not needed for deployment)

Prerequisites

Python 3.8+
CoinGecko API Key (for some endpoints, especially top gainers/losers)
Basic command-line knowledge

Installation

Clone the repository
```
git clone <your-repo-url>
cd ai_agent
```

Create and activate a virtual environment

python -m venv myenv
# On Windows:
myenv\Scripts\activate
# On macOS/Linux:
source myenv/bin/activate

Install dependencies
```
pip install -r requirements.txt
```
Download NLTK data
In a Python shell:
```
import nltk
nltk.download('punkt')
```
Set up your .env file
Create a .env file in the project root:
```
COINGECKO_API_KEY=your_api_key_here
```

Running the App

python app.py

Visit http://127.0.0.1:5000/ in your browser.
Type queries like:
- price of bitcoin
- market cap of ethereum
- trading volume of solana
- top gainers and losers
- ohlc of dogecoin
- predict price of cardano
- Or just say hello!

Example Queries

General:
hello
what is your name?
Crypto Data:
price of bitcoin
market cap of ethereum
trading volume of solana
ohlc of dogecoin
top gainers and losers
recommend a coin
predict price of cardano

Notes

Some endpoints (like top gainers/losers) require a CoinGecko Pro API key.
The price prediction is a simple linear regression and not financial advice.
All environment variables are loaded securely using python-dotenv.

License

This project is for educational purposes. See LICENSE if provided.

rollendxavier/crypto_ai_agent