Published 10 September 2026 · Source dated 3 September 2026
A designer asked an AI agent to audit a Figma button component and found 450 variants across 2,161 layers. Cutting the count to 45 wasn't actually the fix — the real problem was somewhere else entirely.
Story doing the rounds this month: a designer pointed an AI agent at an old Figma button component to see what was actually inside it — 450 variants, 2,161 layers, for one button. First instinct was to slash the variant count (got it down to 45), but on closer look, variant count wasn't the actual problem. The mess was structural, not numerical.
It's a good reminder that AI agents are handy for surfacing what's actually in your design system — not just for generating more of it. There's a growing wave of Figma MCP-focused sessions this September aimed at exactly this: using agents to audit, not just build.
Labelled **adoption**: this is a practitioner's real workflow, not a vendor pitch, though it's still early days for agent-led system audits.
Coach
Open your most-reused component in Figma or Penpot, count the variants, and ask honestly — do you know what each one is for? If not, that's your next cleanup task.
Google's Material Design 3 popularised dynamic, token-based theming back in 2021, and it's still what most design-system setups get measured against. If you're building variables in Figma or tokens in Penpot, M3's structure is worth studying, not because it's new, but because it's proven.
open Material 3's token docs and compare their primitive-vs-role token split against your own Figma or Penpot library — note one gap you could fix this week.
Creative Market's 2026 trend forecast flags 'Glasscore' — blur, glow and translucent layers — as a rising visual style. For teams running token-based design systems, that means more theming flexibility, not just flat colour tokens.
Open your team's (or a starter) design system file and check whether blur/elevation tokens exist — if not, sketch one variant using Figma or Penpot's effect styles.
Some teams are wiring AI coding assistants into their design token workflow, using rules and MCP connections to keep components consistent. It's promising but still fragile in practice.
pick one component in your (or a practice) design system and check whether its tokens are named consistently enough that an AI tool could use them without guessing.
