HN, I recently had to tie out hundreds of PDF documents to an Excel model while on vacation with my Mac. The industry standard for this is a tool called DataSnipper (recently valued at $1B+, costs thousands per seat, and is a heavy Windows-only COM add-in). Running it through Parallels sucks.

I looked for a lightweight, cross-platform alternative and realized there wasn't one. So I built CellSource on nights and weekends. Its been a long time since I was a compsci student so cursor and claude were super helpful.

The goal was to get 80% of the core functionality (snipping, referencing, instant search) without the enterprise bloat. A few technical decisions allow it to be cross-platform.

1) Modern Web Pane vs. COM Add-in: Instead of a heavy local installation, it runs entirely as a modern Office web add-in, which means it works natively on Mac, PC, and Excel for the Web. 2) Metadata over Overlay Shapes: we store the PDF snips and reference links directly in the cell metadata, keeping the workbook incredibly lightweight.

It’s currently live in the standard Excel Add-in store. It is definitely still in beta. If anyone here works with heavy data extraction or just hates bloated enterprise software, I’d love for you to give it a spin and tear apart the UX. Let me know what breaks!

thanks for checking it out