Message Processing Full Flow
- Message Capture: Getting messages through database listening or real-time visual recognition
- Content Analysis: YOLO locate message area + OCR extract text/image information
- Plugin Scheduling: the message goes into the plugin chain for processing (e.g. call OpenAI to generate a reply)
- automatic transmission: By
send_messageInterface simulates manual operation
Development Examples
Inheritance is required to create a pluginPluginClass:
class ReplyPlugin(Plugin):
def process_message(self, msg):
return {"action":"send","content":"已收到您的消息"}
caveat: Contacts with the same name need to use the note name to keep the WeChat window visible in the foreground, the resolution is recommended to be 1080p or above.
This answer comes from the articleOmni-Bot-SDK-OSS: A Visual Recognition-based Automation Framework for WeChat RPAThe
May not be reproduced without permission:AI productivity tools " How to realize the auto-reply of WeChat message by this framework?
