MTG Cardmarket Scanner & Auto-Lister 🃏🤖

A fully automated, end-to-end Python pipeline that takes raw photos of physical Magic: The Gathering cards, identifies them using Computer Vision and OCR, fetches real-time market prices, and can automatically list them for sale on your live Cardmarket (MKM) account.

This project features a "PC Brain + Phone Eye" architecture, allowing you to use your smartphone's camera to seamlessly scan cards directly to your PC over your local network!

✨ Features

• Mobile Camera Integration: Spin up a lightweight local Flask server to turn your smartphone into a wireless scanner. Photos are beamed directly to the bot for instant processing.
• Computer Vision Extraction: Uses OpenCV to automatically detect, crop, and flatten MTG cards from a raw webcam or smartphone photo, regardless of background clutter.
• Smart OCR & Visual Verification: Uses Tesseract OCR with custom noise-reduction post-processing to read card names, and verifies the exact printing/expansion using Scryfall API and ORB feature artwork matching.
• Stealth MKM Scraping: Utilizes nodriver to spin up a stealth Chromium browser, bypassing Cloudflare's bot protection to securely log into your account and scrape live market data.
• Automated Selling (Auto-Lister): Optionally navigates to the Cardmarket Sell page and automatically injects condition, language, and dynamically calculated under-cut pricing to list the card in your live inventory.
• Collection Management: Automatically logs all scanned cards and pricing data (Trend, 30-Day Average, etc.) into a clean CSV database and sorts processed images into /success, /failed, /not_sold directories.

---

🛠️ Prerequisites

1. Tesseract OCR Engine

You must install the Tesseract engine and the required language packs (English, French... ) on your system. Ubuntu / WSL / Debian:

sudo apt update
sudo apt install tesseract-ocr
sudo apt install tesseract-ocr-eng tesseract-ocr-fra

Windows: Download the installer from the UB-Mannheim repository and ensure the executable is in your system PATH.

2. Chromium Browser

nodriver requires a Chromium-based browser to operate its stealth CDP hooks. Ubuntu / WSL / Debian:

sudo apt install chromium-browser

---

🚀 Installation

1. Clone this repository:

git clone [git@github.com:Neyrian/MKM-auto-sell.git](git@github.com:Neyrian/MKM-auto-sell.git)
cd MKM-auto-sell

2. Install the required Python packages:

pip install -r requirements.txt

---

🌐 WSL2 Network Configuration

If you are running this project inside Windows Subsystem for Linux (WSL2), you need to expose the WSL network to your Local Area Network (LAN) so your phone can reach the scanner server.

1. In Windows, open File Explorer, navigate to %USERPROFILE%, and open (or create) a file named .wslconfig.

2. Add the following to enable Mirrored Networking Mode:

[wsl2]
networkingMode=mirrored

3. Save the file. Open an Administrator PowerShell and restart WSL:

wsl --shutdown

4. (Optional) If your Windows Firewall blocks the connection, run this in the Admin PowerShell to allow the scanner port (default 8080) on Private Local Network:

New-NetFirewallHyperVRule -DisplayName "WSL 8080" -Direction Inbound -Action Allow -Protocol TCP -LocalPorts 8080 -Profiles Private -RemoteAddresses 192.168.1.0/255.255.255.0

To remove the rule, use the following command Remove-NetFirewallHyperVRule -DisplayName "WSL 8080"

---

⚙️ Configuration (utils.py)

All core logic, credentials, and toggles are stored in utils.py. Open this file and configure your settings before running the bot.

USERNAME = "YourMkmUsername"     # Your Cardmarket login
PASSWORD = "YourMkmPassword!"    # Your Cardmarket password
USE_MOBILE = False               # Set to True to emulate iPhone, False for Windows Desktop

SELL_CARD = False                # WARNING: Set to True to actually push listings to your live MKM inventory!
FETCH_PRICES = True              # Set to True to launch the browser and fetch MKM prices
DEBUG = True                     # Set to True to save OpenCV debug images

---

📸 Usage (From photos)

1. Create a folder named test (or whatever FOLDER_PATH is set to in utils.py) in the root directory.
2. Drop your raw photos of MTG cards (.jpg, .png) into the test folder.
3. Run the main script:

python mkmbot.py

The Workflow:

1. The script will spin up a stealth browser and log into Cardmarket.
2. If a Cloudflare "Verify you are human" check appears, the script will pause and beep in the console. Complete the captcha manually in the open browser, and the script will automatically resume.
3. It will process images one by one, moving successful scans to /success and failed ones to /failed.
4. Pricing data will be saved to collection_prices.csv.

---

📸 Usage (Mobile Scanning Mode)

1. Start the Bot (Watch Mode): Open a terminal on your PC and run:

python mkmbot.py

Log into Cardmarket. Complete any Cloudflare captchas if prompted. The bot will enter "Watch Mode", waiting for images.

2. Start the Mobile Scanner: Open a second terminal on your PC and run:

python scanner.py

The terminal will output an IP address (e.g., http://192.168.1.11:8080).

3. Scan with Your Phone: (Ensure your phone is on the same Wi-Fi network as your PC.)

   • Open your phone's web browser and go to the IP address displayed in step 2.
   • Tap the 📸 OPEN CAMERA button, snap a picture of the card, and it will instantly beam to your PC. The bot will automatically wake up, process the card, list it, and save the data to your CSV!

📂 Project Structure

• mkmbot.py: The main execution loop and file-sorting logic.
• readingcards.py: Handles all OpenCV image flattening, Tesseract OCR reading, and Scryfall API visual matching.
• authenticate.py: Handles nodriver stealth browser initialization, MKM authentication, price scraping, Cloudflare handling, and auto-listing.
• utils.py: Centralized configuration file for credentials, system toggles, and OpenCV parameters.
• requirements.txt: Python package dependencies.

---

⚠️ Disclaimer & Warning

• Live Account Interaction: If SELL_CARD = True, this script will list items on your real Cardmarket account. Test with cheap bulk commons first to ensure your pricing logic, language mappings, and condition settings behave exactly as you expect.
• Terms of Service: Automated scraping and botting may violate Cardmarket's Terms of Service. Use this tool responsibly, add reasonable delays (asyncio.sleep) between requests, and do not hammer their servers. The developers of this repository are not responsible for banned accounts or incorrect listings.