Visual Workspace for Agents Based on Unix
Hey HN, Thijs here! I'm the founder of Prototyper and today we're launching the first visual canvas built for agents. Couple of interesting lessons from building the product that I think are worth sharing: For the agents, everything is a file. In Prototyper, everything from plans, to apps, and diagrams, can be read as a file. We found that a filesystem is the most natural way for an agent to navigate: it discovers new content and functionality just by traversing the tree. We kept this layer deliberately thin and unopinionated, because it's the substrate, not the experience. It's the…
In plain words
Prototyper is a visual canvas designed for AI agents to plan, build, and organize work. It treats all content—plans, apps, and diagrams—as files in a filesystem structure, allowing agents to naturally discover functionality by traversing the directory tree. Built with a Unix-inspired architecture that works as a distributed, web-based kernel, it integrates with multiple AI models and harnesses without locking users into a specific platform.
written from the facts on this page · September 2026
From the sources
In the maker’s words, at launch
Hey HN, Thijs here! I'm the founder of Prototyper and today we're launching the first visual canvas built for agents. Couple of interesting lessons from building the product that I think are worth sharing: For the agents, everything is a file. In Prototyper, everything from plans, to apps, and diagrams, can be read as a file. We found that a filesystem is the most natural way for an agent to navigate: it discovers new content and functionality just by traversing the tree. We kept this layer deliberately thin and unopinionated, because it's the substrate, not the experience. It's the foundation that makes everything work on top. It essentially operates as a distributed, web-based unix kernel. Because of that it's built for the agents you already use. People run all kinds of models and harnesses, often several at once. With this architecture, we don't try to lock you into one, it's built to work with any agent that you like. That means that any agent can read from and write to your workspace. It's fast. File writes land in under a millisecond. That's not a bogus metric, it's what makes the whole thing feel seamless and like a real extension of your thinking. Thanks to our custom unix kernel. Visual first. Every file in your workspace can be opened on the canvas. Whether you're an engineer working on a new frontend or a PM building a product roadmap. It's a real visual workspace for the actual work, not a description of it. As said, it's not a blank canvas with a box of primitives: the canvas represents a real unix system, which is the kind of purpose-built, opinionated experience that makes a product. In essence, the substrate is generic so the things you build on it don't have to be. What's most interesting is that by this architecture we found that we can decouple system prompt length from agent capability. I'm happy to get feedback from the community and see what you all think of it :).
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, June 2026
the whole month →
Fundraisly▲1,544AI fundraising agent that finds investors and books meetings
AI · Jun 2026 · fundraisly.com
- H6Homebrew 6.0.0▲1,481
Today, I’m proud to announce Homebrew 6.0.0. The most significant changes since 5.1.0 are a new tap trust security mechanism, the new faster, smaller, default internal Homebrew JSON API, sandboxing on Linux, better defaults informed by our user survey, many brew bundle improvements, improved performance and initial support for macOS 27 (Golden Gate). Happy to discuss any questions here!
Dev tools · Jun 2026 · brew.sh
- PU
hope you enjoy
Life & fun · Jun 2026 · vorpus.github.io
- IM
Life & fun · Jun 2026 · hackernewstrends.com