Eunomia
Task organizer with kanban, Eisenhower matrix, node view, and AI photo-to-tasks capture.
Why
I procrastinate. Starting is the hard part — once a task is broken down and I can see it, I do it. The problem is everything before that: overlapping deadlines, half-formed ideas, things I half-remember owing people. It piles up in my head until I freeze.
I tried most of the obvious tools. Todoist, Notion, paper, Things, Apple Reminders. They’re fine for storing tasks, but bad at the part I actually need: looking at everything I have to do and deciding what to do next. None of them made the priorities visible.
So I built Eunomia around an Eisenhower matrix — urgent vs important, four quadrants, every task in one. That’s the killer feature for me. I dump everything onto paper first because that’s still the fastest way to empty my head; then I snap a photo and GPT-4o Vision turns the list into structured tasks I can drag straight into the matrix or a kanban. Paper to system in one move.
I still use Todoist for things with a date attached. Eunomia is for deciding what I should be working on right now.
What it does
- Captures tasks four ways: type them, paste a list, drag and drop, or photograph a paper to-do and let GPT-4o Vision extract the items as structured tasks
- Organizes them in three views: a Kanban board, a node graph for dependency visualization, and an Eisenhower matrix that’s the killer feature — urgent vs important, four quadrants, every task in one
- Encrypts everything client-side with AES-256-GCM via the Web Crypto API; nothing readable ever leaves the device
- Optional team sync: paste the same word sequence on two devices and you share the vault, derived from the words alone with zero server-side knowledge
- Runs as a single static page on Vercel — no build step, no framework, no backend
How it works
Vanilla single-file frontend — no framework, no build step. Data lives in localStorage, encrypted client-side with AES-256-GCM via the Web Crypto API. The encryption key is derived from a user-chosen word sequence, which doubles as the auth credential and as the shared key for optional team sync — paste the sequence on another device and you get the same vault.
Photo-to-tasks uses OpenAI GPT-4o Vision: snap a paper to-do list or whiteboard, the model extracts items and inserts them as tasks automatically.
Hosted as a static site on Vercel.
Status
In active personal use.
See it
- Live: eunomia.top
- Code: private — ask me
- Contact: hello@damianvidalc.com