--- title: Claude Design just got 10X Better... I'm Done | daily.dev description: A prompting technique dubbed the 'gauntlet loop' (also called the design loop), popularized by Matt Shumer and covered by Rubber Nuggets, is presented as a fix... canonical: https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: Claude Design just got 10X Better... I'm Done | daily.dev og:description: A prompting technique dubbed the 'gauntlet loop' (also called the design loop), popularized by Matt Shumer and covered by Rubber Nuggets, is presented as a fix... og:url: https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy og:image: https://api.daily.dev/og/posts/TZZvwsqPY.png og:image:alt: Claude Design just got 10X Better... I'm Done og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # Claude Design just got 10X Better... I'm Done **[YouTube](https://daily.dev/sources/youtube)** · 12 min read · 1 upvotes · 0 comments ## Summary A prompting technique dubbed the 'gauntlet loop' (also called the design loop), popularized by Matt Shumer and covered by Rubber Nuggets, is presented as a fix for Claude's tendency to judge its own design output as good when it isn't. The technique spins up independent 'critic' passes (brief compliance, design system adherence, and craft) that repeatedly challenge Claude's output in a loop until it converges on a polished result. The video walks through building a Claude design system, exporting it, and using a free downloadable 'skill' to automate the critique loop, testing it on a carousel, an HTML animation, and a website mockup. Token costs are discussed (roughly 2-3 million tokens per run across models like Haiku and Sonnet), along with a promoted paid community masterclass. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What is the gauntlet loop technique for improving Claude's design output? The gauntlet loop is a prompting technique that spins up independent 'critic' passes with fresh context to ruthlessly evaluate a Claude-generated design against a brief, a design system, and craft quality, looping until the design reaches an end state. It solves the problem of Claude grading its own work, since asking Claude to critique itself repeatedly yields inconsistent answers. Example runs used three judges across six rounds and consumed roughly two million tokens. _Developers testing new Claude prompting workflows can track techniques like this one via daily.dev._ ### How many tokens does an AI design critique loop typically consume when refining a Claude-generated graphic? A design loop refining an HTML graphic used about 730,000 Haiku tokens and 1.19 million Sonnet tokens across ten critique rounds, while a separate gauntlet loop run with three judges over six rounds consumed around two million tokens, and another example spent three million tokens total. Costs can add up quickly, so this approach is best reserved for high-value, reusable outputs like templates. _Anyone budgeting AI credits for iterative design work can follow token-cost breakdowns like this on daily.dev._ ## Similar posts on daily.dev - [The Anthropic leader who built Claude Code says he ditched prompting — now he just writes loops.](https://daily.dev/posts/the-anthropic-leader-who-built-claude-code-says-he-ditched-prompting-now-he-just-writes-loops--p8qgluhaf) · The New Stack · 1 upvotes · 0 comments --- Tags: [#ai-agents](https://daily.dev/tags/ai-agents), [#ui-design](https://daily.dev/tags/ui-design), [#claude](https://daily.dev/tags/claude), [#prompt-engineering](https://daily.dev/tags/prompt-engineering), [#vibe-coding](https://daily.dev/tags/vibe-coding) [View this post on daily.dev](https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"Claude Design just got 10X Better... I'm Done","url":"https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy"},"datePublished":"2026-08-20T08:09:49.730Z","dateModified":"2026-09-14T07:19:17.454Z","description":"A prompting technique dubbed the 'gauntlet loop' (also called the design loop), popularized by Matt Shumer and covered by Rubber Nuggets, is presented as a fix...","image":"https://i.ytimg.com/vi/jq9LRwE0-GQ/sddefault.jpg","thumbnailUrl":"https://i.ytimg.com/vi/jq9LRwE0-GQ/sddefault.jpg","isAccessibleForFree":true,"articleSection":"YouTube","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"YouTube","logo":"https://media.daily.dev/image/upload/s--W5zMumpP--/f_auto/v1725350181/logos/youtube","url":"https://daily.dev/sources/youtube"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":1},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"ai-agents,ui-design,claude,prompt-engineering,vibe-coding","timeRequired":"PT12M","video":{"@type":"VideoObject","name":"Claude Design just got 10X Better... I'm Done","description":"A prompting technique dubbed the 'gauntlet loop' (also called the design loop), popularized by Matt Shumer and covered by Rubber Nuggets, is presented as a fix...","thumbnailUrl":"https://i.ytimg.com/vi/jq9LRwE0-GQ/sddefault.jpg","uploadDate":"2026-08-20T08:09:49.730Z","duration":"PT12M","url":"https://api.daily.dev/r/TZZvwsqPY","embedUrl":"https://www.youtube.com/embed/jq9LRwE0-GQ"}} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"YouTube","item":"https://daily.dev/sources/youtube"},{"@type":"ListItem","position":3,"name":"Claude Design just got 10X Better... I'm Done"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/claude-design-just-got-10x-better-i-m-done-tzzvwsqpy#faq","mainEntity":[{"@type":"Question","name":"What is the gauntlet loop technique for improving Claude's design output?","acceptedAnswer":{"@type":"Answer","text":"The gauntlet loop is a prompting technique that spins up independent 'critic' passes with fresh context to ruthlessly evaluate a Claude-generated design against a brief, a design system, and craft quality, looping until the design reaches an end state. It solves the problem of Claude grading its own work, since asking Claude to critique itself repeatedly yields inconsistent answers. Example runs used three judges across six rounds and consumed roughly two million tokens. Developers testing new Claude prompting workflows can track techniques like this one via daily.dev."}},{"@type":"Question","name":"How many tokens does an AI design critique loop typically consume when refining a Claude-generated graphic?","acceptedAnswer":{"@type":"Answer","text":"A design loop refining an HTML graphic used about 730,000 Haiku tokens and 1.19 million Sonnet tokens across ten critique rounds, while a separate gauntlet loop run with three judges over six rounds consumed around two million tokens, and another example spent three million tokens total. Costs can add up quickly, so this approach is best reserved for high-value, reusable outputs like templates. Anyone budgeting AI credits for iterative design work can follow token-cost breakdowns like this on daily.dev."}}]} ```