greenyice/rocokingdom_s2_box_detector — explained in plain English
Analysis updated 2026-05-18
Watch a selected screen region and get alerted when a rare box icon appears.
Add your own template screenshots to improve detection for boxes the tool misses.
Adjust resolution presets and matching thresholds to match your game window.
Run the prebuilt Windows executable without installing Python.
| greenyice/rocokingdom_s2_box_detector | 410979729/scope-recall | arahim3/mlx-dspark | |
|---|---|---|---|
| Stars | 33 | 33 | 33 |
| Language | Python | Python | Python |
| Setup difficulty | moderate | moderate | easy |
| Complexity | 2/5 | 3/5 | 3/5 |
| Audience | general | developer | developer |
Figures from each repo's GitHub metadata at analysis time.
README is in Chinese, requires selecting the correct resolution preset and re-selecting the screen region if the game window changes.
This is a screen-reading detection tool for a Chinese online game called Roco Kingdom (also known as Luoke Wangguo). The README is written in Chinese. The tool was created so that players grinding for rare-color box rewards during the game's Season 2 do not have to stare at the screen constantly waiting for specific creatures to appear inside surprise boxes. The program works by capturing a region of the game screen you draw with your mouse, then continuously comparing that region against a library of stored template images using OpenCV, an open-source image processing library. When a match is found that meets a confidence threshold, the tool alerts the player. The author states the recognition accuracy is roughly 85 percent, and notes the limitation is due to an incomplete template dataset. The README invites other players to contribute code or additional template screenshots. A small floating panel appears on screen with buttons to toggle a debug screenshot mode, show a preview of the detection boxes, overlay the detection area directly on the game window, re-select the game region, open settings, and exit. The settings panel lets you adjust matching thresholds, resolution presets from 720p to 4K, frame rate, and the display style of the history log. Several keyboard shortcuts are also available. Adding new templates is possible by taking a screenshot of the target moment and placing it in the templates folder, then registering it through the settings panel. The tool runs entirely locally with no interaction with the game's memory or controls. Installation requires Python with OpenCV, NumPy, PyQt5, mss, and keyboard. A prebuilt executable is provided for users who do not want to run from source.
A screen-watching tool that detects rare creature icons in a Chinese game's surprise boxes using OpenCV template matching.
Mainly Python. The stack also includes Python, OpenCV, NumPy.
The README does not state a license.
Setup difficulty is rated moderate, with roughly 30min to a first successful run.
Mainly general.
This repo across BitVibe Labs
Verify against the repo before relying on details.