[FEAT]: Add ability to retrieve documents from paperless-ngx

[tabbek]

trafficstars

What would you like to see?

I use paperless-ngx ( https://docs.paperless-ngx.com/ ) as a local document store and classification system. It helps me to add metadata and groupings for things like correspondent, document type, tags, etc.

I think it would be useful to be able to have anything-llm have the ability to find and pull documents from paperless-ngx based on things like title, date(s), tags, type, and so on, as context for queries.

[tazzytazzy]

This would be 1000% awesome! Paperless already handles all the OCR and has the text of the PDF's already available along with aging and filenames. :-)

[MrMxyzptlk]

Love this!

[pmffromspace]

Same here

[andrewtcrooks]

This would be amazing!

[thecatontheflat]

Same