Tasker – An open-source desktop agent for browser and OS automation
Hi HN I recently got married, promptly had a bit of a meltdown, and decided to lock myself in a room and build for a while. At the same time, I was trying to outbound sell for my startup and kept running into the same problem: I wanted an automation tool that could actually use my computer like a person. Click through UIs, copy/paste between apps, handle messy workflows — not just APIs and webhooks. I couldn’t find anything that felt: - consumer-friendly (non-technical) - local-first - flexible enough for real-world, UI-driven tasks So I challenged myself to see how far I could get…
In plain words
Tasker is an open-source desktop automation tool that uses AI to perform tasks by interacting with browser and operating system interfaces the way a person would. It handles UI navigation, copy-paste operations between applications, and complex workflows without requiring APIs or webhooks. Designed for non-technical users, it runs locally on desktop computers and works with real-world, messy workflows. The tool is still in early development but has been tested for sales automation and business estimate generation.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hi HN I recently got married, promptly had a bit of a meltdown, and decided to lock myself in a room and build for a while. At the same time, I was trying to outbound sell for my startup and kept running into the same problem: I wanted an automation tool that could actually use my computer like a person. Click through UIs, copy/paste between apps, handle messy workflows — not just APIs and webhooks. I couldn’t find anything that felt: - consumer-friendly (non-technical) - local-first - flexible enough for real-world, UI-driven tasks So I challenged myself to see how far I could get building an open-source, desktop automation app powered by AI. That’s Tasker. I’ve been using it daily for ~2–3 weeks for sales workflows, and my father has been using it to help generate estimates for his HVAC business. It’s still early (still needs to expand to general OS), but it’s already replaced a lot of manual work for us in browser. One thing that’s become very clear: a cloud/deployable version that can run on cron or be triggered via HTTP would unlock a lot of use cases. I’m not totally sure where this goes next, but I wanted to share it early and get feedback. Would love thoughts on: - What workflows you’d actually trust something like this with - Desktop vs cloud tradeoffs - Where this breaks down in practice - Whether this feels useful or just scary Repo and docs are linked on the site.
More ai this month
the category →
I trained a 125M-parameter transformer to autocomplete piano performances in real time (~108 notes/sec on an iPhone 15). The idea is basically GitHub Copilot or Tabnine, except instead of prompting it with code, you prompt it by playing a few notes on a MIDI piano. The model then continues what you played, entirely on-device. The app is free if anyone wants to try it. Happy to answer questions about the model, training, Core ML, or the many things that didn't work.
AI · 17d ago · simedw.com
Astute▲585Automate your B2B brand going viral, with new media creators
AI · 18d ago · company-app.joinastute.com
Hey HN, Henry from Cactus here! We previously released Cactus Needle, a 14MB agentic LLM for tool call, device use, and structured extraction for phones, wearables, smart homes, small robots and microcontrollers. We got really great feedback here, and have now incorporated the suggestions to release Needle 2. The whole model is a single 14MB binary that runs a full session in 28MB of RAM; 45m parameters at 2bit compression. Needle hits 500 tokens/sec decode speed on a Raspberry Pi 5, sits between 400-1,500 tokens/sec on VR devices like Meta Quest 3S and Apple Vision Pro, and ranges…
AI · 26d ago · cactuscompute.com
Launched alongside, January 2026
the whole month →- IN
Hey HN! I wanted to share something I built over the last few weeks: isometric.nyc is a massive isometric pixel art map of NYC, built with nano banana and coding agents. I didn't write a single line of code. Of course no-code doesn't mean no-engineering. This project took a lot more manual labor than I'd hoped! I wrote a deep dive on the workflow and some thoughts about the future of AI coding and creativity: http://cannoneyed.com/projects/isometric-nyc
AI · Jan 2026 · cannoneyed.com
Automatic AI-powered code reviews the moment you open a PR
Dev tools · Jan 2026 · kilo.ai
