In order to meet the needs of different levels of development, Bytebot provides two API interfaces, REST and MCP. The REST API uses the standard HTTP protocol and supports sending commands through the concise JSON format to realize precise control of the mouse, keyboard and screenshots.
MCP (Mouse-Computer-Protocol), on the other hand, is an underlying protocol designed for desktop automation, enabling finer-grained device manipulation. Developers can use these APIs to create complex automation workflows or integrate Bytebot into existing systems.
The API design follows the principle of simplicity and intuition, for example, common operations such as creating tasks and checking status have corresponding endpoints. This dual-API architecture facilitates novices to get started quickly, while meeting the deep customization needs of professional developers, reflecting Bytebot's balance between ease of use and functionality.
This answer comes from the articleBytebot: Automating Desktop Tasks in Linux Containers with Natural LanguageThe
English 
简体中文 
日本語 
Deutsch 
Português do Brasil