AI tools like Claude Code have made it possible for users to build software with no coding knowledge whatsoever. That’s not to say the process is easy, though: You may not need to write code directly, but you need to understand how your computer’s terminal works, how to deploy and maintain software, and deal with lots of other associated tasks. Raycast, the launcher app that has been particularly popular among Mac users, thinks it can make the process even simpler. The company is launching a new product called Glaze that attempts to make it easy to build, use, share, and discover new vibe-coded software. Right now it’s only available for Mac, but Raycast plans to bring it to Windows and mobile over time, and the company thinks it could change the way you think about your apps.
“Glaze is our take on personal computing,” says Raycast cofounder Thomas Paul Mann. He loves the idea of letting users build tiny utilities for themselves, or hyper-specific apps to fit their team’s hyper-specific needs. You can use Glaze to build whatever you want, or browse the directory of apps made and shared by others. Or, better yet, Mann says, grab someone else’s app and then tweak it to your exact liking.
The Glaze process is even more straightforward than most vibe coding tools: you just type a prompt, and the tool tries to create an app in one go. Claude Code and OpenAI’s Codex are the platform’s primary underlying models, so the build process might feel familiar to existing vibe coders — some upfront questions and a few checks along the way — but so far in my testing, I’ve found that Glaze tries extra hard to finish the job the first time.
Mann confirms this is the goal: “We want to make sure you can just prompt anything you want,” he says. “If you have to dive into the code, we basically did something wrong.” Glaze is meant to take care of things like cloud storage, to follow basic tenets of good design, and to manage any necessary APIs and integrations. These are features most users take for granted in software, but require real knowledge and effort to build, even in Claude Code. Glaze tries to wave them all away again.
Over Zoom, Mann shows me a bunch of apps he’s made. There’s one that generates an emoji from any picture you select. There’s a simple spending tracker. There’s one for recording Zoom meetings and highlighting key moments. There are data-viz dashboards and project trackers and tweet analyzers and logo creators and lots more. Each one is built very simply and follows all of Apple’s Liquid Glass rules, has a somewhat retro-styled skeuomorphic icon, and lives in a list called “My Projects” inside the Glaze app.
Glaze is not technically a Raycast feature, but it is deeply integrated with the launcher. When you build Glaze apps, Mann says, “you can think about it coming with a bundled extension, which Raycast can pick up and make its own.” Raycast’s job in this case is to orchestrate things across apps, help you find the ones you’re looking for, and act as the launcher it already is.
The idea here starts simple, but gets complicated fast. When you build an app with Glaze, and someone else installs it from the Glaze Store, are they installing your app, or is it more like they’re downloading some code from GitHub to run for themselves? Put another way, who is responsible for the app working correctly? Can someone else add a feature to your app and call it their own? Can you charge money for your app? If something goes wrong, who’s to blame?
Mann says he doesn’t know all the answers. He’s not even sure exactly what to charge for Glaze, or who’s most likely to pay for it. (So far, the plan is to have a free version, and then a few $20-$30 paid tiers based on usage.) Nobody knows those things, really; this whole software ecosystem is so new that nobody has figured out how it’s supposed to work. So far, Mann’s theory is that most people will want to build simple, largely single-player tools that run locally on their machine. Glaze doesn’t seem to aspire to be the home of the next big social network or the next Salesforce, just a lot of little ways to make those things better.
But that’s not to say Raycast’s ambitions aren’t huge. They are: Mann tells me he thinks we’re at “the iTunes moment” for software, when suddenly everything you want can be available in a single place. “I think it’s a fundamental change in software,” he says. Over time, he sees this kind of prompted app changing the app economy entirely. “In some regard, we’re taking on the App Store on Mac and Windows,” Mann says. “And who knows? Maybe we can take them over.”
