Design platform Figma announced a suite of new features at its first Config event in India, headlined by a Markdown-based editor designed to help companies define design standards and steer AI agents toward predictable outcomes. The company unveiled Figma Specs, a tool that lets teams document their design language in a simple format that both human designers and AI agents can reference. The move comes as Figma works to address AI's non-deterministic nature by giving companies more control over how automated tools adhere to brand rules.
Figma Specs functions as a Markdown editor where users can define design language in a straightforward way, either as a standalone file or embedded within the canvas. Beyond text editing, the tool allows teams to reference other Figma elements, embed FigJam diagrams, code layers, and designs, and pull in design libraries, code libraries, and guidelines to create a centralized context for the design system. Agents can consult this centralized system to follow established rules. The company also moved its code layers feature, which adds coding to designs on canvas for interactive prototypes, from private beta in June to public beta this week. The platform now supports over 20 new tools through its MCP, including Open Code, Dia, Grok Bot, and Indian AI startup Sarvam. Additionally, Figma is integrating Payload, the open-source headless CMS it acquired last year, with its Figma Make product for generating dynamic websites and managing content directly; Payload's downloads have climbed sixfold in the past 12 months.
According to Figma's chief design officer Loredana Crisan, the platform sees itself as "the place where software teams come to decide what they build and ship." Crisan explained that AI has changed what's available, noting that "there are 17-times more lines of code being produced today than in the past. But when you're looking at releases, only 1.3-times more releases." Much of that code, she said, isn't necessarily built to ship but to discover what the product is—essentially built to design. The company stated that India is its largest market by user count outside the United States and that it works with 40% of the top 100 companies listed on the Bombay Stock Exchange.
The features reflect Figma's effort to balance AI's creative potential with the need for consistency and brand control. The company made its agent on canvas available to a wider user base in the past week, but the new Specs editor and design libraries provide guardrails to steer AI toward more predictable results. Users can now ask the agent to convert a design into code or modify parts of the code to iterate on a prototype, streamlining the path from concept to interactive testing. By centralizing design language, code libraries, and brand guidelines in one referenceable system, Figma aims to prevent AI-generated work from veering off-brand while still allowing teams to move quickly.
The company's expansion in India—where it opened its first office late last year with sales, partnership, go-to-market, and product support functions—signals confidence in the market's design community. The integration of Payload and the public rollout of code layers suggest Figma is positioning itself not just as a design tool but as a full production environment where teams can prototype, code, and publish without leaving the platform. For organizations navigating AI adoption, the ability to define design rules in a simple, centralized format could become critical as agents take on more creative tasks. The stakes extend beyond workflow efficiency: as design teams generate exponentially more code without a proportional increase in shipped releases, tools that help distinguish exploration from execution will likely shape how companies build software in an AI-saturated environment.