PDF Workbench

Local, browser-based PDF tools built on Streamlit and PyMuPDF. Files never leave your machine.

Viewer: read PDFs, ctrl/cmd + scroll to zoom.
Editor: click anywhere on a page to add text, then download the result.
Workbench: pick pages with thumbnails, search and filters, then merge, save a selection, split by ranges / every N pages / selected pages, or extract text, tables, formulas and images to a zip.
Organizer: drag pages across PDFs into a new order, rotate, duplicate, delete, then build one PDF.

Run

uv sync
uv run streamlit run app.py

uv run pytest

app.py sets up navigation and the sidebar uploader.
pages/ holds one script per screen.
src/pdf_workbench/ holds the PDF operations (basic_ops, split, organizer, text_editor, extract) and the Streamlit controls that drive them (*_controls.py, page_selector.py).
page_grid_frontend/ and text_editor_frontend/ are the two custom components, plain HTML/CSS/JS with no build step.