Design systems tighten up as agents start editing files
Published 10 September 2026 · Source dated 10 September 2026
As AI agents get write access to design files, teams are doubling down on strict token naming and component governance so automated edits don't quietly break consistency. It's a shift from "nice to have" documentation to enforced structure.
With agents now able to generate or remix designs directly in-file, design system teams are treating naming conventions, token structure and component documentation as guardrails, not paperwork. A messy library that "sort of works" for humans breaks fast when an agent is matching against it programmatically. The fix teams are reaching for: stricter naming, fewer one-off overrides, and documented rules an agent (or a new hire) can actually follow. This isn't a flashy trend — it's the boring governance work getting more attention because the cost of skipping it just went up. Established practice, newly urgent.
Coach
Pick one component in your Figma or Penpot library and check its naming and token structure against your team's docs — fix the gaps before an agent (or a teammate) trips over them.
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.
