A little open sourced present for the LinkedIn Community: a workflow
to annotate and organize your Desktop Screenshots, aka to obsidianize them. Link to the repository 👇
The workflow allows you to automatically annotate all the screenshots laying down on your desktop, which let’s be honest, if you build products and are also constantly communicating about them, can pile up quickly.
It uses ollama locally and Gemma4 self hosted, so you don’t give away sensitive information you may not want to. Steps to have that workflow running are included in the repository.
Annotation is fairly simple, but you can fork the repository or submit a PR with changes and adapt it to what is most convenient for you.
Running a python command will move from this👇
<img src="https://media.daily.dev/image/upload/s--vkS-9XKP--/f_auto/v1776692842/ugc/content_aa69df44-4102-4c5e-9b17-ab0da0cd7719?_a=BAMAMiWQ0" alt="Screenshot 2026-04-20 at 15.47.17.png">
to this👇
<img src="https://media.daily.dev/image/upload/s--xWWMwRPI--/f_auto/v1776693225/ugc/content_b1e52719-57d1-4320-a55d-aadf4a10adee?_a=BAMAMiWQ0" alt="Screenshot 2026-04-20 at 15.53.39.png">
Link to repository 👇
<a href="https://github.com/ndr3svt/desktop-screenshot-vault" target="_blank" rel="noopener nofollow">https://github.com/ndr3svt/desktop-screenshot-vault</a>

Andres Villa Torres (ndr3svt)

An open-source Python workflow that automatically annotates desktop screenshots and organizes them into an Obsidian vault. It runs locally using Ollama with Gemma4, keeping sensitive information private. A single Python command processes scattered desktop screenshots into structured, annotated notes in Obsidian.

An Open Source Workflow to Annotate in Markdown your Desktop Screenshots Clutter