Figma announces new features to define how agents work on its platform
During its inaugural Config event in India, design powerhouse Figma unveiled a suite of new capabilities aimed at bridging the gap between human creativity and machine-generated output. By introducing a Markdown-based editor and deeper integrations with the Payload headless content management system, the company is providing teams with the tools to establish strict design guardrails for AI agents.
Establishing the "Source of Truth" with Figma Specs
As AI agents become more prevalent in the design workflow, the challenge of maintaining brand consistency has grown. To address this, Figma has debuted Figma Specs, a sophisticated Markdown editor designed to act as a centralized reference point for both human designers and AI agents.
Figma Specs allows teams to document their design language, guidelines, and system rules in a format that is easily readable by machines. These documents can exist as standalone files or be embedded directly within the design canvas. Key features of the new tool include:
- Cross-referencing: Users can link directly to specific elements on the canvas.
- Rich Embedding: The editor supports embedded FigJam diagrams, code layers, and design assets.
- Contextual Management: Teams can pull in design and code libraries to ensure that AI agents have the necessary context to adhere to brand standards.
By providing this structured "definition" of a brand, Figma aims to steer AI toward more predictable, on-brand outcomes, mitigating the inherent non-deterministic nature of generative models.
Code Layers Move to Public Beta
Beyond design documentation, Figma is doubling down on the intersection of design and development. The company has officially moved its code layers feature into public beta. First teased in a private beta this past June, this functionality allows designers to inject actual code into their canvas elements, enabling the creation of highly interactive, functional prototypes.
Users can now leverage AI agents to transform static designs into code or iterate on existing prototypes by modifying code snippets directly within the platform. This shift reflects Figma’s broader ambition to be the primary workspace where software teams decide what to build and ship.
“We see Figma as the place where software teams come to decide what they build and ship. And so for that, I think one of the key things to keep in mind about what AI has changed and what it made available is the ability to build,” said Loredana Crisan, Figma’s Chief Design Officer.
Crisan noted a significant industry trend: while code production has surged 17-fold, actual product releases have only increased by 1.3 times. Figma’s latest updates are designed to help teams bridge this gap, ensuring that the code being written is truly aligned with the product vision.
Expanding the Ecosystem
Figma is also aggressively expanding its interoperability. The platform now supports over 20 new tools via its Model Context Protocol (MCP), including Open Code, Dia, Grok Bot, and the Indian AI startup Sarvam.
Furthermore, the company is deepening its integration with Payload, the open-source headless CMS it acquired last year. By connecting Payload with Figma Make, users can now generate dynamic websites while managing content directly within the design environment. This integration comes at a time of significant growth for Payload, which has seen its download volume increase sixfold over the past year.
A Strategic Focus on India
The timing of these announcements at the Config event in India underscores the region's importance to Figma’s global strategy. Since opening its first office in the country late last year, India has become Figma’s largest market outside of the United States. With the platform currently utilized by 40% of the top 100 companies listed on the Bombay Stock Exchange, Figma is clearly positioning itself as the backbone of India’s rapidly evolving digital product landscape.