- PDF: extracts selectable text via pymupdf, falls back to Tesseract OCR for scanned docs
- PDF: renders first page as screenshot thumbnail
- Images: Tesseract OCR for text extraction, OpenAI vision API fallback for photos
- Plain text files: direct decode
- All extracted text stored in extracted_text field for search/embedding
- Tested: PDF upload → text extracted → AI classified → searchable
New deps: pymupdf, pytesseract, Pillow
System dep: tesseract-ocr added to both Dockerfiles
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
Full backend service with:
- FastAPI REST API with CRUD, search, reprocess endpoints
- PostgreSQL + pgvector for items and semantic search
- Redis + RQ for background job processing
- Meilisearch for fast keyword/filter search
- Browserless/Chrome for JS rendering and screenshots
- OpenAI structured output for AI classification
- Local file storage with S3-ready abstraction
- Gateway auth via X-Gateway-User-Id header
- Own docker-compose stack (6 containers)
Classification: fixed folders (Home/Family/Work/Travel/Knowledge/Faith/Projects)
and fixed tags (28 predefined). AI assigns exactly 1 folder, 2-3 tags, title,
summary, and confidence score per item.
Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
