Skip to content

getting-started.md

Latest commit

 

History

History
212 lines (143 loc) · 4.14 KB

getting-started.md

File metadata and controls

212 lines (143 loc) · 4.14 KB

Getting Started with Polymarket AI Bot

This guide will get you up and running in 10 minutes.

Prerequisites

  • Python 3.10 or higher
  • Polymarket account (optional for testing)
  • Anthropic API key (Get one here)

Step 1: Installation

# Navigate to the project directory
cd probablyprofit

# Install dependencies
pip install -r requirements.txt

Step 2: Configuration

Copy the example environment file:

cp .env.example .env

Edit .env and add your API keys:

# Required - at least one AI provider
ANTHROPIC_API_KEY=sk-ant-...
# or OPENAI_API_KEY=sk-...
# or GOOGLE_API_KEY=...

# Optional (for live trading) - your Polygon wallet private key
PRIVATE_KEY=0x_your_private_key_here

# Optional (for news context)
PERPLEXITY_API_KEY=pplx-...

Step 3: Run the Quickstart

Test that everything works:

python quickstart.py

This will:

  1. Connect to Polymarket
  2. Fetch current markets
  3. Ask Claude to analyze them
  4. Show you the AI's reasoning

Step 4: Run an Example Bot

Try one of the three example strategies:

Momentum Bot

python examples/momentum_bot.py

Contrarian Bot

python examples/contrarian_bot.py

News-Driven Bot

python examples/news_bot.py

Step 5: Create Your Own Bot

Create a new file my_bot.py:

import asyncio
import os
from dotenv import load_dotenv
from probablyprofit import AnthropicAgent, PolymarketClient, RiskManager

load_dotenv()

# Define your strategy in natural language!
MY_STRATEGY = """
You are a [YOUR STRATEGY HERE] trader.

Your strategy:
1. [STEP 1]
2. [STEP 2]

Entry rules:
- [RULE 1]
- [RULE 2]

Exit rules:
- Take profit at [X%]
- Stop loss at [Y%]
"""

async def main():
    client = PolymarketClient(
        private_key=os.getenv("PRIVATE_KEY"),
    )

    risk_manager = RiskManager(initial_capital=1000.0)

    agent = AnthropicAgent(
        client=client,
        risk_manager=risk_manager,
        strategy_prompt=MY_STRATEGY,
    )

    await agent.run()

if __name__ == "__main__":
    asyncio.run(main())

Run it:

python my_bot.py

What's Next?

1. Learn Prompt Engineering

Read strategy-guide.md to learn how to write effective trading strategies.

2. Explore the API

Check out api-reference.md for detailed API documentation.

3. Backtest Your Strategy

Before risking real money, backtest via CLI:

probablyprofit backtest --strategy "your strategy" --days 30

Or check backtesting results in the web dashboard.

4. Use Paper Trading

Test with fake money first:

# Via CLI
probablyprofit run "your strategy" --paper --paper-capital 10000

# Or via Python
agent = AnthropicAgent(..., paper_trading=True)

5. Go Live (Carefully!)

When you're ready:

  1. Start with small position sizes
  2. Monitor closely for the first few days
  3. Gradually increase capital as you gain confidence

Common Issues

"Missing API key"

Make sure .env has your ANTHROPIC_API_KEY.

"Cannot place order"

If you get this error, it means you're running in read-only mode (no Polymarket credentials). This is fine for testing - the bot will still analyze markets, it just won't execute trades.

"Rate limit exceeded"

You're making too many API calls. Increase loop_interval in your agent configuration.

Safety Tips

⚠️ Important:

  1. Start small - Use tiny position sizes initially
  2. Paper trade first - Test thoroughly before using real money
  3. Monitor closely - Check your bot frequently at first
  4. Set strict limits - Use conservative risk management settings
  5. Understand the code - Read through the framework before using it
  6. Only risk what you can afford to lose

Getting Help

  • 📖 Read the docs in docs/
  • 💬 Check GitHub issues
  • 🐛 Report bugs on GitHub

Summary

You now have:

  • ✅ A working AI trading bot framework
  • ✅ Three example strategies to learn from
  • ✅ Tools to create your own custom bots
  • ✅ Risk management and backtesting capabilities

The power is in the strategy prompt - experiment, test, and refine!

Happy trading! 🚀