How to Build a Python Bot to Automate a Simple Web Game

Introduction

Many gamers are familiar with cheats, but this article shows how to create a simple game bot using Python. The goal is not to gain unfair advantage in large online games, but to practice programming techniques by automating a small browser game where customers order sushi.

Tools Preparation

Install the autopy, PIL (Pillow), and pywin32 packages. autopy provides cross‑platform mouse and keyboard simulation and screen capture. PIL handles image processing, and pywin32 simplifies some Windows‑specific actions.

For screen capture you can use any tool; the author uses PicPick. An editor such as Vim is used for coding.

Mouse Movement

The command autopy.mouse.move(x, y) moves the cursor to the specified screen coordinates (origin at the top‑left corner). Note that the library’s coordinate conversion may be slightly inaccurate due to integer division: point.x *= 0xFFFF / GetSystemMetrics(SM_CXSCREEN); A more precise formula would be:

point.x = point.x * 0xFFFF / GetSystemMetrics(SM_CXSCREEN);

Mouse Click

Click actions are fast; insert a short time.sleep() if the game does not respond quickly enough.

Keyboard Operations

Keyboard input is not required for this particular game, so it is omitted.

Image Capture and Analysis

Capture the game screen with PIL.ImageGrab.grab() or autopy.bitmap.capture_screen(). Determine the fixed positions of customer avatars and ingredient slots relative to the game window.

Similar Image Search Principle

Instead of using autopy.bitmap.find_bitmap (which requires exact pixel matches), the tutorial implements a custom image fingerprinting method. Each ingredient image is resized to a small fixed size (e.g., 18×13) and converted to a binary hash.

The similarity between hashes is measured using Hamming distance:

def hamming_dist(self, hash1, hash2):

return sum(itertools.imap(operator.ne, hash1, hash2))

If the minimum distance is below a threshold (e.g., 50), the corresponding dish is identified; otherwise the region is considered empty.

Automated Cooking

Store the recipe for each of the eight dishes as a list of ingredient coordinates. The bot moves the mouse to the ingredient locations, clicks them, and drags the finished dish to the customer. The following class encapsulates the process (code omitted for brevity): # class definition image omitted Overall, the bot demonstrates basic screen‑based automation, image processing, and simple pattern matching, providing a hands‑on example for improving Python programming skills.