Kalshi Sports Bot (Simplified)

A lightweight Python toolkit for experimenting with Kalshi prediction markets. It provides a small Flask dashboard, a command-line menu, and helper clients for Kalshi and The Odds API. The decision engine relies on free odds data, free ESPN context, and a heuristic scorer—no paid AI services are required.

Features

• Interactive CLI for scanning markets, launching the dashboard, and checking portfolio status.
• Flask dashboard with JSON endpoints for bot control, portfolio insights, and trade helpers.
• Odds and context ingestion from The Odds API and ESPN, combined with heuristic scoring to suggest trades.
• Optional auto-trading flow (disabled by default) that requires valid Kalshi credentials.

Prerequisites

• Python 3.9+
• API keys (set via environment variables or a local .env file):
  • KALSHI_API_KEY and KALSHI_API_SECRET for authenticated Kalshi trading (required for real orders and portfolio data).
  • ODDS_API_KEY for The Odds API (required for scanning odds).

Installation

1. Clone the repository and move into it:

   git clone <repo-url>
   cd ProjectBroke

2. Create and activate a virtual environment:

   python -m venv .venv
   source .venv/bin/activate  # Windows: .venv\Scripts\activate

3. Install dependencies:

   pip install -r requirements.txt

Configuration

Environment variables can be loaded from a .env file in the project root. At minimum, set your API keys:

KALSHI_API_KEY=your_key
KALSHI_API_SECRET=your_secret
ODDS_API_KEY=your_odds_api_key

Project layout

• main.py – command-line entry point with an interactive menu and subcommands for scanning, dashboard launch, and portfolio checks.
• web/app.py – Flask dashboard exposing JSON endpoints for bot status, manual scans, and basic trading helpers.
• src/decision_engine.py – builds betting recommendations from odds data and heuristic scoring.
• src/kalshi_client.py – Kalshi REST client with RSA signing support.
• src/sports_odds_client.py – wrapper for The Odds API odds and event retrieval.

Running the CLI

All commands load environment variables from .env if present.

• Interactive menu: python main.py
• Launch dashboard server: python main.py dashboard
• Scan a sport: python main.py scan americanfootball_nfl --max 5
• View portfolio balances (requires Kalshi credentials): python main.py portfolio
• Check which API keys are configured: python main.py status

Dashboard

Start the Flask dashboard with python main.py dashboard (or run web/app.py directly). It uses the same environment variables as the CLI. The dashboard keeps a rolling cache of decisions, portfolio data, and recent trades while the background scanner iterates through the enabled sports list.

Key API routes:

• Bot control: /api/bot/status, /api/bot/start, /api/bot/stop, /api/bot/config (GET/POST), /api/bot/auto-trade (POST), /api/bot/scan-now (POST)
• Portfolio & history: /api/portfolio, /api/positions, /api/trades
• Decisions & logs: /api/decisions, /api/debug/logs, /api/debug/settings
• Trading helpers: /api/trade/place (POST), /api/trade/cancel/<order_id> (DELETE), /api/orders
• Health & stats: /api/health, /api/stats, /api/errors

Scanning uses The Odds API for price snapshots plus optional free ESPN data. Recommendations are produced by a simple heuristic scorer (no paid AI). Auto-trading is disabled by default; toggling it only updates in-memory state—you must still provide valid Kalshi credentials for real orders.

Development and testing

Run the Flask app directly during development:

python web/app.py

If you add tests under tests/ or web/, you can run them with:

python -m pytest

Notes and limitations

• The decision engine uses heuristic scoring based on odds edges and bookmaker coverage.
• API clients print errors to stdout but do not include retry/backoff beyond basic rate-limit handling.
• This code is for experimentation and should be used responsibly—understand the risks before trading.

License

MIT License. See LICENSE for details.