About agentic-doc-ext

are there any agent or node in nodered that can extract any text from the documents ?

What sort of document?
There are a few HTML scraping nodes for example.

invoices and scanned bills and invoices

This node seems like it will do what you describe: