CL-SN55 941 —
GPT-61S 885 -0.6%
GPT-56T 861 —
GPT-6S 825 -1.4%
CL-OP55X 819 -1.4%
MUSE-SPK 793 -1.4%
QWEN-38X 779 -2%
GPT-6A 778 -0.5%
CL-OP5H 712 -1.7%
CL-OP5X 710 -1.7%
KIMI-K3X 700 -2.5%
CL-OP46H 695 -1.7%
CL-FAB5H 693 -1.7%
CL-OP55H 693 -1.7%
CL-OP47H 684 -1.7%
GEM-37FH 649 -1.8%
GEM-38FH 634 -3.4%
GPT-56S 574 -0.5%
CL-OP47 540 -0.6%
INKL 531 —
GEM-3P 494 —
CL-OP46 490 —
CL-OP48 483 —
GEM-31P 459 -5.4%
GPT-6L 436 —
CL-SN55 941 —
GPT-61S 885 -0.6%
GPT-56T 861 —
GPT-6S 825 -1.4%
CL-OP55X 819 -1.4%
MUSE-SPK 793 -1.4%
QWEN-38X 779 -2%
GPT-6A 778 -0.5%
CL-OP5H 712 -1.7%
CL-OP5X 710 -1.7%
KIMI-K3X 700 -2.5%
CL-OP46H 695 -1.7%
CL-FAB5H 693 -1.7%
CL-OP55H 693 -1.7%
CL-OP47H 684 -1.7%
GEM-37FH 649 -1.8%
GEM-38FH 634 -3.4%
GPT-56S 574 -0.5%
CL-OP47 540 -0.6%
INKL 531 —
GEM-3P 494 —
CL-OP46 490 —
CL-OP48 483 —
GEM-31P 459 -5.4%
GPT-6L 436 —
← Back to feed

Anthropic Ships Claude Design Overhaul: System Imports, Code Round-Trips, Token Fix

Anthropic has shipped a significant update to Claude Design, the conversational prototyping tool that launched in April as a research preview and generated strong early traction. Three major gaps from the initial release are now addressed.

Design system imports. Teams can load existing Figma tokens, component libraries, and design system definitions directly into Claude Design sessions. Previously, users had to re-specify brand rules — colours, spacing, typography, component behaviour — in natural language at the start of each session. Now those constraints come from the source of truth: the design system itself. For teams with mature design systems, this is the difference between Claude Design as a prototyping novelty and Claude Design as a repeatable workflow tool.

Bidirectional code round-trips. The update closes the loop between visual output and generated code. Edits to the code now reflect visually; changes to the visual prototype propagate to the underlying code. This was the critical missing feature for engineering teams: without bidirectional sync, Claude Design produced throwaway prototypes rather than artifacts that could connect to actual development work.

Token efficiency fix. The most common early complaint was cost unpredictability. Small iterative changes — adjusting padding, tweaking a colour, rearranging components — triggered full context re-processing on each edit, making iterative design sessions expensive and hard to budget. The update patches context handling so incremental edits process as deltas rather than triggering wholesale revalidation of the full session context.

Why the Token Fix Matters

Design work is inherently iterative. A product designer might make dozens of small adjustments in a single session. If each adjustment carries the token cost of starting fresh, the economics fall apart quickly and teams stop using the tool for real work. The token-burning problem was not marginal — it was structural, and fixing it required rethinking how Claude Design tracks state across edits. Until that was resolved, professional teams were effectively limited to demos rather than production use.

Context

Claude Design launched in April as a research preview positioned as a competitive response to Figma’s own AI prototyping features. Figma’s stock dropped 7% on the announcement. The original launch generated immediate developer interest, but the token-burning problem and the absence of design system integration kept it from being practical for production workflows.

The three-feature overhaul addresses exactly those gaps: import your existing systems instead of re-describing them, maintain code that stays in sync rather than diverging after the first edit, and pay for actual changes rather than full-context restarts.

Anthropic has not announced pricing changes alongside this update.