Skip to main content

ITmatterss

Figma adds new tools to guide AI design agents: Here’s how it works

Vertical Share Bar
Figma adds new tools to guide AI design agents

News in Short

  • Figma has introduced Specs, a Markdown-based editor for documenting design rules and brand guidelines.
  • The new tools aim to help AI agents follow design systems more consistently.
  • Code layers are now available in public beta for interactive prototypes.
  • Figma is expanding Model Context Protocol support to more than 20 tools.
  • The company is integrating Payload with Figma Make for dynamic website creation and content management.

Figma has introduced Figma Specs, a Markdown-based editor that helps teams document their design language and guide AI agents. The company announced the feature at its first Config event in India on October 14.

The editor lets teams describe design rules in a format that both people and AI agents can reference. It can exist as a standalone file or alongside a design canvas.

Users can also reference Figma elements, embed FigJam diagrams and include code layers and designs. This gives teams a central place to explain how their products should look and behave.

The move addresses a growing challenge in AI-assisted design. AI agents can generate interfaces quickly, but their output may not consistently follow a company’s established visual identity. Figma Specs aims to provide clearer instructions for those workflows.

Figma adds design and code libraries for AI agents

Figma is expanding the context available to AI agents through design libraries, code libraries and design system guidelines. These resources can help agents understand a company’s existing components and rules before creating or modifying interfaces.

The goal is to make AI-generated designs more predictable and consistent with a brand’s established system. Teams can use the shared references to guide both Figma’s own agent and supported third-party agents.

The update comes as Figma expands access to its agent on the canvas. Since AI-generated results can vary, the company is adding controls that help users steer agents towards more consistent outcomes.

These tools could be particularly useful for teams working across multiple products or designers. They can establish common rules without requiring every instruction to be repeated manually.

Figma code layers enter public beta

Figma has also moved its code layers feature into public beta, opening it to more users for testing. The feature brings code into the design canvas and allows teams to build more interactive prototypes.

Users can ask an AI agent to turn a design into code or modify specific sections of existing code. This creates a workflow in which designers can experiment with functionality alongside visual layouts.

Figma first announced code layers in a private beta in June 2026. The public beta expands access to the feature as the company develops its AI-powered design and prototyping tools.

Loredana Crisan, Figma’s chief design officer, told TechCrunch that AI has significantly increased the volume of code being produced. She cited figures suggesting that code output has increased 17-fold, while releases have grown only 1.3 times.

Her point highlights a distinction between generating code and shipping products. Much of the additional code can help teams explore ideas and discover what they want to build before committing to a final product.

Figma expands MCP support and integrates Payload

Figma is adding support for more than 20 tools through its Model Context Protocol (MCP) integration. The newly supported tools include Open Code, Dia, Grok Bot and Sarvam, an Indian AI startup.

MCP support helps connect AI tools with external context and capabilities. The expansion gives teams more options for incorporating Figma into their AI-assisted development workflows.

The company is also integrating Payload, the open-source headless content management system it acquired last year, with Figma Make. The integration is designed to support dynamic website generation and direct content management.

Figma Make lets users generate websites, while Payload can manage the content behind them. Bringing the two together could help teams move from website concepts towards content-driven implementations within a more connected workflow.

Figma says Payload downloads have increased sixfold over the past 12 months, indicating growing adoption of the project.

Why Figma’s India event matters

The announcements reflect Figma’s broader effort to connect design, coding and website development within one platform. Rather than focusing only on generating visual layouts, the company is adding tools that help teams guide AI agents and turn prototypes into more functional experiences.

India is an important market for this strategy. Figma opened its first Indian office in late 2025, covering sales, partnerships, go-to-market and product support.

The company says India is its largest market by users outside the United States. It also works with 40% of the top 100 companies listed on the Bombay Stock Exchange.

With Specs, code layers, expanded MCP support and Payload integration, Figma is building more structure around AI-assisted product development. The challenge will be ensuring these tools help teams produce reliable, usable software rather than simply generating more designs and code.

24

Leave a Reply

Your email address will not be published. Required fields are marked *

logo

Get the latest news instantly

You can change your preferences anytime.