CS2 Marketplace Arbitrage Analyzer

A TypeScript tool that identifies profitable arbitrage opportunities for Counter-Strike 2 (CS2) items between the Steam Community Market and Skinport marketplace.

Overview

This tool analyzes price differentials between two major CS2 item marketplaces to find profitable trading opportunities. It evaluates both inbound (Skinport -> Steam) and outbound (Steam -> Skinport) strategies, accounting for marketplace fees and 7-day trade hold periods.

Trading Strategies

Inbound (Buy on Skinport, Sell on Steam)

• Target: Items priced significantly below Steam buy orders
• Fee consideration: 15% Steam marketplace fee
• Requires: 50+ Steam sales in past 7 days for reliable data

Outbound (Buy on Steam, Sell on Skinport)

• Target: Items where Skinport prices are close to Steam sell prices
• Fee consideration: 6-12% Skinport fee (6% for items GBP800+)
• Requires: 25+ Skinport sales in past 7 days for reliable data

Features

• Fetches real-time price data from Steam and Skinport APIs
• Calculates 7-day price statistics (avg, min, max, volume, std dev)
• Computes profitability ratings based on configurable thresholds
• Analyzes order book depth with closeness coefficient
• Exports results to CSV for further analysis
• Rate limit handling with automatic retry logic

Item Categories

The tool supports multiple CS2 item categories:

• Cases (40+ different cases)
• Knives (Karambit, Bayonet, Bowie Knife with various finishes)
• Rifles (AK-47, AWP, M4A1-S, M4A4, and more)
• Heavy weapons (MAG-7, Nova, Sawed-Off, XM1014, M249, Negev)

Requirements

• Bun runtime (v1.0+)
• Steam account with Community Market access
• Skinport API credentials

Setup

1. Clone the repository:

git clone https://github.com/yourusername/steam-arbitrage-bot.git
cd steam-arbitrage-bot

2. Install dependencies:

bun install

3. Create environment file:

cp .env.example .env

4. Configure your .env with API credentials (see .env.example for details)

Usage

Run analysis for a specific category:

bun run start:case    # Analyze CS2 cases
bun run start:knife   # Analyze knives
bun run start:rifle   # Analyze rifles
bun run start:heavy   # Analyze heavy weapons

Results are saved to ./output/{date}/{category}/ with:

• raw_data.csv - Complete data for all items
• inbound.csv - Top 10 inbound opportunities
• outbound.csv - Top 10 outbound opportunities

Project Structure

src/
├── config.ts              # Item catalogs and fee constants
├── type.d.ts              # TypeScript type definitions
├── run.ts                 # Core orchestration logic
├── steam.repository.ts    # Steam Market API client
├── skinport.repository.ts # Skinport API client
├── rating.util.ts         # Profitability rating calculations
├── rating.constant.ts     # Rating thresholds
├── price.util.ts          # Price statistics calculations
├── order.util.ts          # Order book analysis
├── result.util.ts         # CSV export and data transformation
├── number.ts              # Number formatting utility
├── runs/                  # Category-specific entry points
│   ├── case.ts
│   ├── knife.ts
│   ├── rifle.ts
│   └── heavy.ts
└── data/
    └── itemId.json        # Pre-fetched Steam item ID cache

Development

bun run lint      # Run ESLint
bun run format    # Format code with Prettier
bun run test      # Run tests

Rating System

Inbound Ratings

Items are rated based on profit potential (buy price vs Steam buy order):

• 1.0: 70%+ profit potential
• 0.8: 60-70% profit potential
• 0.6: 50-60% profit potential
• 0.4: 48-50% profit potential

Outbound Ratings

Items are rated based on expected loss (Steam sell vs Skinport average):

• 1.0: 5% or less loss
• 0.8: 5-10% loss
• 0.6: 10-13% loss
• 0.4: 13-15% loss

Marketplace Fees

• Steam: 15% seller fee
• Skinport: 12% fee (6% for items GBP800+)

Limitations

• Steam session credentials expire and need periodic refresh
• API rate limits may require staggered execution
• 7-day trade hold affects arbitrage timing
• Price volatility during hold period creates risk

License

MIT