Claude Code's New /design Command, Explained
Anthropic added a /design research preview to Claude Code in the week of 17–21 August 2026, letting the coding agent draft editable UI artboards for you.
Anthropic added a /design command to Claude Code as a research preview during the week of 17–21 August 2026, according to the tool's official "What's new" changelog [1]. The command brings what Anthropic calls "Claude Design's artboard workflow" into both the command-line tool and Claude Code Desktop: a developer runs /design with a brief, and Claude publishes a canvas of editable artboards for a UI design, which the developer can pick from, tweak, and then have Claude implement directly in code [1]. Anthropic states it requires Claude Code version 2.1.233 or later and is available on Pro, Max, Team and Enterprise plans [1].
The feature shipped as part of a wider release spanning versions 2.1.234 to 2.1.239, alongside two other additions covered below: a new "Concise" output style, and mobile access to remote coding sessions [1].
What Anthropic actually announced
Anthropic's own example shows the intended workflow: a developer types a prompt such as "redesign the composer based on what people actually use it for," and Claude responds by publishing a link to a canvas of design options [1]. The developer opens that canvas, selects an artboard, and tells Claude which one to implement — at which point Claude writes the corresponding code [1]. Anthropic describes the underlying mechanism as built on "artifacts," the same publishing system Claude uses elsewhere to turn a session's output into a shareable, live page [1].
Why this is a different kind of feature for a coding tool
Coding assistants, including Claude Code itself until this point, have worked almost entirely in text and code — reading files, writing diffs, running commands. /design instead asks Claude to produce and present visual, selectable design options before any code is written, closer to how a designer might mock up several directions before a developer picks one to build. Anthropic labels it a "research preview," the same designation it has used for other Claude Code features — such as /ultrareview and computer use — before they moved to general availability.
Two companion features in the same release
Alongside /design, Anthropic introduced a "Concise" output style in version 2.1.237, which the changelog describes as making Claude "lead with the result and skip preamble and narration, while doing the work as thoroughly as in the Default style" [1]. Anthropic specifies that error reports, security warnings and confirmations for destructive actions retain their full content regardless of this setting, and that Claude gives full detail again whenever a user explicitly asks for an explanation [1]. Separately, the changelog confirms that Claude Code's "Remote Control" feature — letting a phone start or check on a coding session running on another machine — has come out of research preview, with connected machines now appearing as device cards in the Claude mobile app [1].
Why this matters
Most AI coding tools split cleanly into "writes code" or "generates images/designs," rarely both. By having Claude draft multiple visual UI options and then implement whichever one is chosen, Anthropic is testing whether a single coding agent can absorb part of the design step too — potentially useful for small teams or solo developers who don't have a dedicated designer to hand off to. That it launched as a research preview, alongside a Concise output style aimed at reducing unnecessary verbosity, also points to Anthropic tuning Claude Code toward faster, more autonomous day-to-day use rather than only adding raw capability.
Who should care
Solo developers and small teams building product interfaces without in-house design support are the most direct beneficiaries, since /design offers an option between hand-coding a UI from scratch and hiring a designer for every iteration. Existing Claude Code users on Pro, Max, Team or Enterprise plans already have access to try it, provided they're on version 2.1.233 or later. Teams that have adopted a stricter, more concise AI-output style for their engineers may also want to evaluate the new Concise mode specifically, independent of the design feature.
Practical implications for buyers and users
Because /design is a research preview, teams should treat it as an experimental workflow to try on a real but low-stakes UI task first, rather than something to route production design decisions through immediately. The feature depends on Claude's "artifacts" publishing system, so teams should check their plan's artifact availability before assuming /design will work identically across every account type. For engineering managers evaluating Claude Code against other coding assistants, the combination of /design, Concise output, and general-availability Remote Control in one release is a reasonable signal of how frequently Anthropic is shipping meaningful changes to the tool — roughly weekly, based on the versioned "Week 34" cadence Anthropic uses for its own changelog.
Limitations, availability and unresolved questions
As a research preview, /design may change materially or be withdrawn before wider release, and Anthropic has not published a timeline for when — or whether — it will reach general availability. The changelog does not specify how design revisions are handled if a developer wants changes beyond picking one of the presented artboards, nor how the feature performs on more complex, multi-screen design tasks versus the single-component example Anthropic shows. It's also not stated whether /design usage is metered or priced differently from Claude Code's standard usage.
Verdict
/design is a genuinely novel addition for a coding assistant — most competitors remain text- and code-only — and pairing it with a Concise output style and general availability for Remote Control makes this a substantial weekly release rather than a single-feature update. Its research-preview status is an honest signal that Anthropic itself doesn't consider it finished, and teams should treat early results accordingly. For developers without dedicated design support, it's worth trying on a real task; for teams with an established design process, it's more likely to be a curiosity than a replacement for now.
The current AI Writing & Research shortlist
Where this sits in the wider market: our current shortlist for AI Writing & Research, what each tool is best at and the main caution to check before committing.
| Tool | Best for | Current position | Important caution |
|---|---|---|---|
| ChatGPT Best all-rounder | General writing, analysis and multimodal work | GPT-5.6 combines strong reasoning with files, images, tools and broad workflow support. It is the safest starting point when one assistant must cover many jobs. | Teams should define data-handling rules and verify important claims. |
| Claude Long-form pick | Editorial work, complex documents and careful reasoning | Claude’s current Opus and Sonnet 5 family is built for sustained professional and agentic work, with a strong reputation for readable long-form output. | The highest-capability tiers can be unnecessary for routine copy. |
| Gemini Google ecosystem | Workspace users and multimodal source material | Gemini 3.7 Flash, documented in August 2026, is the current Flash release, connecting reasoning, multimodal inputs and Google’s productivity ecosystem. | Feature availability varies by Workspace plan and region. |
| Perplexity Research pick | Fast web research and cited discovery | Perplexity is useful when the first requirement is finding and comparing live web sources rather than drafting from memory. | A citation does not guarantee that the source supports every sentence; open the evidence. |
| Jasper Brand governance | Marketing teams with repeatable brand workflows | Jasper focuses on governed marketing content, brand context and campaign production rather than being a general-purpose chatbot. | Plans, limits and model availability change frequently; confirm the current vendor page before purchasing. |
| Copy.ai GTM workflows | Sales and marketing process automation | Copy.ai has evolved from a copy generator into a go-to-market workflow platform for repeatable content and sales operations. | Plans, limits and model availability change frequently; confirm the current vendor page before purchasing. |
| Writesonic AI visibility | SEO content and answer-engine monitoring | Writesonic combines assisted content production with tooling aimed at search and AI-answer visibility. | Plans, limits and model availability change frequently; confirm the current vendor page before purchasing. |
| Grammarly Editing layer | Everyday rewriting, tone and quality control | Grammarly works best as an editing and communication layer across existing applications rather than as the only writing system. | Plans, limits and model availability change frequently; confirm the current vendor page before purchasing. |
| Notion AI Knowledge workspace | Teams whose documents and projects already live in Notion | Notion AI is strongest when it can work inside an existing team knowledge base instead of requiring constant copying between tools. | Plans, limits and model availability change frequently; confirm the current vendor page before purchasing. |
| KoalaWriter SEO drafts | Structured long-form drafts and niche publishing | KoalaWriter remains a focused option for producing structured, search-aware drafts quickly. | Human research, original experience and fact-checking are still required before publishing. |
Related reading
Sources and verification notes
Primary product documentation checked for this update: