see what you
actually worked
on today.
privacy-first activity tracking for knowledge workers.
local models. intelligent digests. no cloud required.
capture
every 60 seconds, workloom checks your active window. smart filter rules decide what to record — skipping spotify, flagging youtube, logging your code editor.
analyze
a local vision model reads the screenshot. it extracts what you were doing, any visible text, and classifies the activity type. nothing leaves your machine.
digest
at 10 pm, sessions are assembled into a structured report. distractions, insights, ideas worth revisiting. saved as markdown to your local drive.
local first
screenshots never leave your machine. a local vision model reads them, extracts what matters, then they're gone. only text summaries are stored.
daily digests
every evening, a structured report appears. what you worked on, what pulled you off track, ideas that surfaced. all in markdown. all yours.
interactive chat
ask your activity history anything. find the article from tuesday. understand what triggered your context switch. powered by semantic search.
fully customizable
change filtering rules, digest structure, and summary tone through markdown files. no code required. your workflow, your language.
open source
built in python. runs on mac, windows, and linux. no subscriptions, no accounts, no tracking. fork it, extend it, make it yours.
your data,
your machine.
screenshots are analyzed in memory and never written to disk. window titles and summaries stay local. the only thing that touches the cloud is your final digest — and only if you choose a cloud provider. you can run entirely offline with a local ollama model.
built for the way knowledge workers actually work.
GET STARTEDopen source · python 3.11+ · mac · windows · linux