Anthropic has unveiled a new feature for its AI-powered development platform, Claude Code, introducing the /design skill aimed at streamlining user interface (UI) design workflows. Currently in research preview, this addition enables developers to generate and refine UI artboards directly within the Claude Code environment.
Traditionally, UI design involves multiple tools and transitions between design software, written specifications, and coding environments. The /design skill integrates these steps by allowing developers to prompt Claude to create interface concepts, select and modify layouts, and implement the finalized designs—all within a single platform.
Built upon Anthropic’s Artifacts framework, which facilitates the creation and editing of interactive outputs, the /design skill allows users to invoke the feature by entering /design in the command-line interface or desktop application. Claude then generates visual interface drafts that are editable, enabling developers to adjust the proposed UI before converting it into functional code.
For instance, a developer working on a security operations dashboard can specify requirements such as an alert queue, severity filters, an incident timeline, an Indicator of Compromise (IOC) search panel, and analyst notes. The /design skill would generate multiple dashboard layouts as editable artboards. The developer can choose a preferred structure, make necessary revisions, and instruct Claude to implement the selected interface.
This integration signifies a shift in AI-assisted development, moving beyond code generation to encompass the entire design-to-implementation process. By bridging the gap between natural-language product requirements, visual prototyping, and code generation, Claude Code offers a more cohesive development experience.
However, developers should exercise caution when adopting AI-assisted UI development. It’s essential to review generated code for security vulnerabilities, validate accessibility, and ensure the application meets all functional requirements before deployment. Treating AI-generated front-end code as untrusted until thoroughly reviewed is a prudent approach to maintain application security and integrity.
The /design skill is available to Claude Pro, Max, Team, and Enterprise subscribers. Users may need to update Claude Code to access this new feature.
As AI continues to evolve, integrating design and development workflows into a single platform could significantly enhance productivity and innovation. However, the success of such integrations will depend on the quality of the generated designs, the ease of customization, and the ability to produce maintainable, secure code ready for production.
