GreenYIce

GreenYIce / RocoKingdom_S2_Box_Detector

Public

为了在洛克王国S2赛季刷异色盒子时，不用一直盯着，所以写了这个小工具。感谢deepseek，感谢gpt，感谢gemini，感谢cc。基于OpenCV实现，准确度80%左右，可以实现检测惊喜盒子的中层和下层

14
3
85% credibility
Found May 28, 2026 at 14 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

This is a screen-capture based detection tool for the game Roco Kingdom. It watches a user-selected region of the game, uses multi-frame analysis to reliably spot items and stats (called bloodlines and attributes), and displays results in a floating overlay panel. The tool runs in the background while you play, snapping pictures of your screen and comparing them against template images to identify game elements. It includes both a detailed recognition history view and a simpler screenshot counting mode, plus a full settings panel for customizing detection behavior. The overall code quality is solid with proper threading, caching, and a comprehensive PyQt5 interface.

How It Works

1
🎮 Download a gaming helper tool

You hear about a tool that helps track items and stats in Roco Kingdom by watching your game screen automatically.

2
🖥️ Select where you play

A simple startup screen asks you to pick your game resolution and whether you're playing solo or with a partner, then confirms your choice.

3
📦 Draw a box around the game area

You drag your mouse across the entire game window to tell the tool exactly where to watch for action.

4
👀 Watch the magic unfold

The tool silently watches your screen, snapping multiple quick pictures and combining their results to reliably spot what you're looking for.

5
📝 See your recognition history

Each detected item appears in a floating panel on your screen, with colored chips showing bloodline and attribute results, plus a running count of what you've seen.

6
Customize how it works
🔍
Recognition Mode

View detailed history with color-coded results and vote counts for each detected item.

📷
Screenshot Mode

Simply count how many boxes you've clicked through, showing the latest captures side by side.

✨ Track your progress effortlessly

Your item history builds up automatically while you play, giving you instant visual feedback on everything the tool recognizes.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 14 to 14 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is RocoKingdom_S2_Box_Detector?

This is a Python screen detection tool for the game Roco Kingdom that automatically spots rare "surprise boxes" during S2 season farming. Instead of staring at the screen, you draw a region over the game window and the tool watches for specific patterns using OpenCV template matching. Results appear in a floating overlay with match history and statistics. It supports both one-handed and two-handed play modes.

Why is it gaining traction?

The project fills a real automation gap for players grinding the S2 season. Unlike generic image recognition libraries, this tool is purpose-built for one game's interface with presets for 720p through 4K resolutions. The detection uses multi-frame voting and blur filtering to reduce false positives. A full settings panel lets you adjust thresholds, add new templates, and tweak ROI positions without editing code.

Who should use this?

Roco Kingdom players doing S2 season box farming who want to run the game unattended. The GUI makes it accessible to non-programmers. Developers curious about OpenCV-based game automation might find the cascade detection approach useful as a reference implementation.

Verdict

The 0.8500000238418579% credibility score signals a very early-stage project with 14 stars and minimal documentation. The code structure is solid and the feature set is surprisingly complete for a personal tool, but it has not been battle-tested by a community. Download and try it if you play the game. Evaluate it as a working prototype, not production software.

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.