EDITION OF OCTOBER 15, 2026

Figma wants to give teams and AI agents shared rules

Figma is announcing a Markdown editor to formalize design rules and provide AI agents with more context. The company is also expanding prototyping capabilities and integrating Payload into Figma Make. These announcements aim to provide clearer guidance for creation, without yet demonstrating its reliability.

Figma wants to give teams and AI agents shared rules — A symbolic composition about a model and its connections.
A symbolic composition about a model and its connections.DailySynth illustration

Figma is seeking to make design guidelines more accessible to AI agents. With Specs, teams can formalize their visual language in a reference document. The aim is to guide generation and prototyping, but the reported announcements do not yet allow reliability gains to be measured.

Figma is announcing new tools to clarify the rules its agents and those from third-party providers must follow when producing interfaces. According to TechCrunch, these additions were presented at the company's first Config event in India. Their stated goal is to improve adherence to brands' visual identities during AI-assisted creation.

The main addition, Figma Specs, is a Markdown editor designed to describe an organization's design language. The document can serve as a reference for both teams and agents. It can exist as a standalone file or sit within the Figma canvas. Beyond text, it allows users to reference canvas elements and embed FigJam diagrams, code layers and graphic designs.

Figma also plans to centralize more context around the design system: graphics libraries, code libraries and guidelines. Agents can consult this collection to guide their work. The potential benefit lies in the consistency of the references: a rule spelled out in a shared space could limit divergent interpretations among the people and tools contributing to the same project.

Figma wants to give teams and AI agents shared rules — A conceptual diagram connecting data, a model and its uses, without numerical results.
A conceptual diagram connecting data, a model and its uses, without numerical results.DailySynth illustration

This approach accompanies broader access to the agent within the canvas. TechCrunch highlights AI's nondeterministic nature and presents these tools as a way to guide results toward greater predictability. This describes a product intention. The document provides no measurements that would establish whether agents actually follow a brand's rules more closely.

Prototyping is also evolving as code layers enter public beta. This feature adds code to designs on the canvas to make prototypes more interactive. Users can ask the agent to turn a design into code or modify a portion of the code to develop the prototype further. The beta status remains essential when assessing this expansion to more users.

Finally, Figma is integrating the Payload content management system into Figma Make to generate dynamic websites and manage their content directly. These announcements thus bring visual design specifications, prototyping and editorial management closer together. Their practical impact will, however, depend on how agents use the references provided and on teams' ability to spot deviations before accepting a proposal.

In this edition.

The next edition, at the right time.

Follow upcoming editions ↗