Published 24 July 2026 · Source dated 1 March 2024
The Design Tokens Community Group's shared token format is quietly becoming what AI design agents rely on to output on-brand results. It's been in progress since 2019 — infrastructure, not a trend.
AI agents generating screens on canvas (Figma) or importing designs (Penpot) are only as good as the design system feeding them. The Design Tokens Community Group (DTCG) has spent years defining a shared, tool-agnostic token format so colour, spacing and type mean the same thing across Figma, Penpot, and code.
**Why established, not hype**: this group has published drafts since 2019 — slow, deliberate standards work, not a launch. For juniors, the practical upshot: if your design system's tokens follow (or move toward) DTCG format, AI agents and cross-tool exports behave far more predictably. If they don't, that's a real gap worth flagging to your lead.
Coach
Open your current design system's token file and check whether names/structure are consistent — flag any mess you find to your lead this week.
Figma argues MCP servers only work well when a real design system backs them — otherwise AI drifts off-brand. This is adoption-stage: teams with strong systems already benefit, but it depends entirely on your system's maturity.
figma.com ·
Established
Your design system is becoming an AI's reading list
Figma's Dev Mode MCP server exposes variables, component maps and Code Connect mappings directly to coding agents like Copilot, Cursor and Claude Code, instead of a flat screenshot. That means how you name and structure tokens now directly affects what an AI can build correctly.
figma.com ·
Established
Dev Mode MCP grounds AI code in your system
Figma's Dev Mode MCP Server feeds your actual design system — components, tokens, styles — into AI coding agents like Cursor and Claude Code, so generated code matches what your team already built. It's been live over a year and is now routine, not a novelty.