8000 GitHub - blacksyncai/arbitrage-bot: Multi-platform prediction market arbitrage scanner — Polymarket, Kalshi, Gemini Predictions with Claude AI verification · GitHub
[go: up one dir, main page]

Skip to content

About

Multi-platform prediction market arbitrage scanner — Polymarket, Kalshi, Gemini Predictions with Claude AI verification

Resources

Stars

2 stars

Watchers

0 watching

Forks

Latest commit

 

History

1 Commit

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Arbitrage Bot

A multi-platform prediction market arbitrage scanner that monitors Polymarket, Kalshi, and Gemini Predictions simultaneously, using Claude AI to verify every opportunity before alerting.

How It Works

  1. Fetches active markets from all 3 platforms every 30 seconds
  2. Matches markets across platforms using NLP similarity scoring
  3. Calculates net profit after all platform fees
  4. Sends every potential opportunity to Claude AI for verification (prevents false positives)
  5. Alerts on verified opportunities above the profit threshold

Platforms Covered

Platform Markets Taker Fee Maker Fee
Polymarket 200+ 0.75–1.80% (varies by category) 0%
Kalshi 99+ ~2.0% ~0.1%
Gemini Predictions 500+ ~1.75% ~0.5%

Setup

1. Install dependencies

pip install requests anthropic python-dotenv rich

2. Configure API keys

cp .env.example .env
# Edit .env and fill in your keys

You need at minimum an Anthropic API key for Claude verification:

For live trading (optional), you also need keys for each platform:

3. Run the bot

python run_bot.py

The bot runs in dry run mode by default — it scans and alerts but does not place orders. To enable live trading, set dry_run=False in run_bot.py and ensure your platform API keys are configured.

Architecture

arbitrage_bot/
├── run_bot.py              # Main entry point and scan loop
├── bot/
│   ├── polymarket_client.py  # Polymarket API client
│   ├── kalshi_client.py      # Kalshi API client  
│   ├── gemini_client.py      # Gemini Predictions API client
│   ├── matcher.py            # NLP-based cross-platform market matching
│   ├── detector.py           # Arbitrage opportunity detection + fee math
│   ├── scanner.py            # Multi-platform scan orchestration
│   ├── claude_verifier.py    # Claude AI opportunity verification layer
│   └── models.py             # Shared data models
└── .env.example              # API key template

Fee Structure & ROI

The bot calculates net profit after all fees. The minimum threshold is configurable in run_bot.py:

MIN_PROFIT_PCT = 0.005   # 0.5% minimum net profit
ALERT_THRESHOLD = 0.02   # 2%+ triggers a loud alert

Real arbitrage windows typically appear during:

  • Breaking news events (one platform reprices faster than another)
  • New market listings (price discovery period)
  • Large whale trades temporarily moving one platform's price

Disclaimer

This is a research tool. Prediction market trading involves financial risk. Always verify opportunities manually before executing trades.

About

Multi-platform prediction market arbitrage scanner — Polymarket, Kalshi, Gemini Predictions with Claude AI verification

Resources

Stars

2 stars

Watchers

0 watching

Forks

Releases

Packages

Contributors

Languages

0