OCR read and ai to rename and categorize

[neonnightsnz]

OCR and chatgpt could be used to read pdf docs and give them a more appropriate name and category

files would be scanned and pop up approving the new name and category manna adjustment and corrected. Data from this processing would be used to create a sorting models

https://nanonets.com/ http://www.dropitproject.com/

[tfeldmann]

Hi, thanks for the ticket. This is possible right now with a python filter and the confirm module. Something like this:

rules:
  - locations: .
    filters:
      - python: |
          # AI help here:
          import random
          num = random.randint(1, 1000)

          return {
            "ai_name": f"new-name-{num}.txt"
          }
    actions:
      - confirm: "Should this file be renamed to {python.ai_name}?"
      - echo: "Renaming..."

[tfeldmann]

Two things already on the todo list:

• Something like a prompt action instead of the confirm action would be more fitting here, so the AI-generated name could be corrected.
• A way to write plugins which can be distributed via pypi. I guess a lot of people would love to use a helper like this.

[staberas]

I'm currently working integrating ai features, currently implemented ai template generator that will generate a yaml when you tell it what you want to do. All that using local llm. https://github.com/staberas/AI-organize