ChromaScape is a colour-based automation framework designed for Old School RuneScape. Inspired by Kelltom's OSBC, SRL-dev, Nick Cemenenkoff's RuneDark and SlyAutomation, it focuses on education, iterative prototyping, human like interaction using solely pixel based logic.
Whether you're just starting out or building advanced automation systems, ChromaScape provides a modular, structured framework to help you create robust bots and learn by doing.
Check out a full step by step guide on how to get started Read the Installation Guide
- Please visit the Wiki for detailed guides on writing scripts with this framework.
- Feel free to join the Discord
ChromaScape provides a layered architecture with tightly scoped utilities that each serve a specific role.
Due to the separation of domain, core and actions utilities it provides a greater level of control and expansion, the nature of these utilities is detailed in the wiki.
Here's a very simple scripting example from the DemoMiningScript that you'll see working below.
private void clickOre() {
try {
BufferedImage gameView = controller().zones().getGameView();
Point clickLoc = PointSelector.getRandomPointInColour(gameView, "Cyan", 15);
if (clickLoc == null) {
stop();
return;
}
controller().mouse().moveTo(clickLoc, "medium");
controller().mouse().leftClick();
} catch (Exception e) {
logger.error(e);
logger.error(e.getStackTrace());
}
}Recording2.mp4
ChromaScape uses advanced remote input techniques to function as a virtual second mouse dedicated to the client window. Unlike traditional input methods, this approach never hijacks your physical mouse, so you can continue using your PC without interruption while the bot runs in the background.
This is achieved through KInput.
ChromaScape uses a slightly modified version of the 64-bit supported version of KInput. The original KInput source is also available for reference.
There are instructions on how to build KInput from source within the third_party directory in DEV_README.md
To further reduce bot detection risks, ChromaScape uses an adapted version of WindMouse, a physics based calculation of gravity and wind to ensure pixel imperfections unlike bezier curves. WindMouse has been a successful staple within the community for over a decade, and provides exceedingly human mouse movements.
Recording3.mp4
The UI is built with Spring Boot, a mature industry framework, and served locally. This gives you a powerful way to view logs, manage scripts, see the bot's sensor information, and extend functionality. It's fully customizable with basic HTML/CSS/JS, so power users can tweak or overhaul it without modifying core framework utilities or needing to worry about tight coupling.
Newly: you can now have runelite covered by other applications while the bot runs
You can't minimise it or put it entirely offscreen yet, but you can place other applications on top and have it run in the background.
This tool allows you to pick exact pixel colours directly from the screen. It’s useful for identifying precise colour values needed to detect specific game elements or interface components. The colour picker supports real-time sampling and stores these colours for use in detection routines. Advanced users may choose to instantiate and store colours in code, per their preference. Inspired by ThatOneGuy's BCD
Note: You will need all other colours except your desired one to be black.
Using the colours obtained from the picker, the framework scans defined screen regions to find matching outlines or clusters of pixels. This process enables the bot to recognise NPCs, static objects and or RuneLite's indicators by their unique colour signatures. The detection logic is optimized to handle slight variations in colour due to lighting or graphical effects by allowing for a lower and upper range of HSV colours.
ChromaScape includes functionality for identifying images or sprites by comparing pixel patterns against stored templates. This technique allows detection of complex objects like UI elements, sprites, items, and RuneLite plugin additions.
ChromaScape utilises template matching for accurate and fast OCR. This solution - as opposed to machine learning - provides for ocr at runtime. This was inspired by SRL and OSBC.
This project downloads specific fonts and UI elements from the OSBC, RuneDark and SRL-dev projects to enable accurate template matching, OCR, and UI consistency. These resources are used solely for educational and research purposes and they are not packaged directly within this repository.
