Skip to content
UsecaseBoard
← All use cases
githublife

Suggest document metadata and improved OCR inside a Paperless archive

At a glance

Paperless-GPT adds model-assisted metadata and optional OCR processing to paperless-ngx, with manual and automatic tags and documented non-destructive testing options. This workflow is listed under life, document-organization, home-admin.

Agents / tools
paperless-gpt, paperless-ngx, Ollama or hosted LLM, OCR

Building a tool for these workflows? See weekly rates.

Save ideas in this browser. Upvotes count across the board.

Prompt excerpt

Follow the repository's setup with a test paperless-ngx archive and your chosen model provider. Start with manual processing and the documented non-destructive PDF mode. Review proposed titles, tags, correspondents, and OCR text against the originals before enabling limited automatic processing of selected document tags.
Explore all →
hackernewslife

Extract tagged Paperless receipts into schema-defined JSON

The receipthero-ng author describes detecting tagged Paperless documents and converting receipts or invoices to usable JSON using schema definitions, with Docker Compose setup documented in the linked project.

receipthero-ngpaperless-ngxDocker Compose+1
Source ↗