Mon, the design Grok Bot
Design and build any screen
Bloom, the design system Grok Bot
Builds or audits your design tokens
Lens, the UI review Grok Bot
Reviews a screen like a senior designer
North, the brand Grok Bot
Name, voice, wordmark and palette
Board, the dashboard Grok Bot
The operating view, with every state
Frame, the mobile app Grok Bot
App screens at 375 by 812, no phone chrome
Gate, the paywall Grok Bot
States the limit and the reason to pay
Till, the checkout Grok Bot
Checkout that does not lose the sale in the last four fields
Key, the authentication Grok Bot
Sign in, sign up, reset, every error
A11y, the accessibility Grok Bot
Contrast, focus, labels, keyboard. Fixes, not a lecture
Grok Bots for design work: building a landing page, a dashboard, a mobile app screen, a paywall, a checkout or a sign in flow, setting up a design system, and reviewing what is already there. Each one is a full instruction set, so the Bot returns a screen you can run rather than advice about layout.
The reason a design Grok Bot is worth more than a general model asked for a landing page is that the standard is written down. One type family and four sizes. One ink, two greys and a single accent that belongs to the primary action. Proof before features. Real images rather than a grey box where a picture belongs. A general model averages what it has read, and the average of the web is a nine icon grid with a gradient headline.
The other difference is that these return something that runs. Mon writes into your project in your framework, Board builds the dashboard with every state including the panel that failed, Frame draws the app screen at 375 by 812 with no phone frame around it, and A11y hands back corrected markup rather than a list of rule numbers. If you have a codebase, point them at it and they work in your components and your tokens.
Mon for when you need a whole screen built and do not want to start from an empty file
Pitch for when the page has to sell, and the argument matters more than the layout
Board for when a team is running the business from a spreadsheet somebody updates by hand
Frame for when it ships in an app and has to work under a thumb
Lens for when the screen exists and something is wrong that nobody can name
Bloom for when the same blue is defined in nine places across the codebase
A11y for when it has to pass an audit rather than a taste test
Key for when sign in, sign up and every error state have to be built once, properly
Design and build the screen. Mon reads your project, matches your framework and your tokens and writes the files. Board builds a dashboard, Frame builds mobile app screens at 375 by 812, Till builds checkout. They return something that runs, not a description of a layout.
Mon if you are building something and do not know what it should look like. Lens if you already built it and want it reviewed. Bloom if the same colour exists in nine places in your codebase. A11y if it has to pass an accessibility audit.
No. Tell it what you are making in a sentence and it decides the structure, writes the real copy and builds it. The standards are written into the instructions, which is the part you are getting.
They are included in the BrowserMoon licence, which is one payment for every design template and every Bot template, for unlimited projects including work you sell.
A design Grok Bot writes into the project you already have, so the design templates for the tool you build in are the other half of the same job: Lovable templates, Bolt templates, Replit templates, Claude Code templates, ChatGPT Codex templates, Cursor templates, Windsurf templates, Framer templates, v0 templates and Cline templates.