Figma Ships Specs and Code Layers Beta to Tame AI Agents
At its first Config India event, Figma launched Figma Specs, a Markdown editor defining design rules for AI agents, moved code layers to public beta, and expanded MCP support to 20+ tools.
Updated
Why it matters
- Figma announced Figma Specs, a Markdown editor for defining design rules that AI agents must follow, at its first Config event in India.
- Figma's chief design officer Loredana Crisan said 17x more lines of code are produced today versus the past, while releases grew only 1.3x.
- Code layers moved from private beta (announced June) to public beta at Config India this week.
- Figma now supports over 20 new tools via MCP, including Open Code, Dia, Grok Bot, and Sarvam.
- Figma says it works with 40% of India's top 100 Bombay Stock Exchange companies; Payload downloads grew sixfold in 12 months.
Figma announced a Markdown-based editor called Figma Specs at its first Config event in India, aiming to give companies a way to define how AI agents work on its platform. The feature lets users write down a brand's design rules and have agents — and humans — follow them.
Figma Specs is a plain Markdown editor that describes a company's design language. The document can sit as a standalone file in Figma or co-exist on the canvas. Beyond text, users can reference other Figma elements: canvas objects, FigJam diagrams, code layers, and designs. Teams can also attach design libraries, code libraries, and guidelines, creating a centralized context that agents must consult before they generate anything.
The timing matters. Figma made its on-canvas agent available to a wider user base in the past week, and the company is candid about why it now needs guardrails. Because AI is non-deterministic — the same prompt can produce different outputs — Figma is shipping tools to steer it toward more predictable outcomes. In plain words: Figma is writing down the rules because it cannot guarantee the AI will follow them on its own.
What else shipped at Config India?
Figma moved its code layers feature from private beta, announced in June, into public beta. The feature adds code directly to designs on canvas, producing more interactive prototypes. Users can ask the agent to turn a design into code, or edit part of the code, to iterate on a prototype.
The company also said it now supports more than 20 new tools through MCP (Model Context Protocol, an open standard that lets AI applications connect to external data sources and services). The list includes Open Code, Dia, Grok Bot, and the Indian AI startup Sarvam. The takeaway: Figma is positioning itself as the hub other AI tools plug into, not just a destination for its own agent.
What did Figma's design chief say about AI-generated code?
Chief Design Officer Loredana Crisan gave TechCrunch the sharpest statistic of the event: "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." Her argument is that most AI-generated code is not meant to ship at all. "A lot of that code actually is not necessarily built to ship, but really built to discover what the product is. It's built to design."
She framed the stakes for Figma directly: "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."
The 17x-versus-1.3x gap is the company's core claim, and it is worth noting what it does not say: Figma did not publish the source or methodology behind those figures. Still, the gap supports Figma's strategic bet — if AI inflates code volume without matching release velocity, the bottleneck shifts toward deciding what to build, which is Figma's territory.
How does the Payload acquisition fit in?
Figma acquired the open-source headless CMS (a content management system that delivers content via API rather than rendered pages) Payload last year. At Config India, the company said it is integrating Payload with Figma Make, its product for generating dynamic websites and managing content directly.
Figma reported that Payload's downloads have increased sixfold in the last 12 months. The company did not disclose absolute download numbers, so the growth rate is hard to independently weigh. The move signals Figma wants to own the full path from design to live, content-driven site.
Why India, and why now?
Figma opened its first India office late last year, housing sales, partnership, go-to-market, and product support functions. The company said India is its biggest market by users outside the United States, and that the platform works with 40% of India's top 100 companies listed on the Bombay Stock Exchange.
Holding Config in India for the first time, adding Sarvam to its MCP partners, and citing BSE-listed clients all point one direction: Figma is treating India as a primary growth market, not a satellite one. Whether Specs and the expanded agent tooling can keep enterprise design systems coherent as AI generation scales will be the real test of this release cycle.
Original: figma.com
More from James Calloway
Show full bio
News editor covering people, launches and day-to-day practice in artificial intelligence and machine learning at AI In Context.
276 articles
Related articles
- OpenAI Codex and Figma Launch Two-Way Code-to-Design Workflow
- Shopify Launches Canvas, an AI Chat Interface for Building Stores
- Microsoft Rebuilds Copilot App Around Office, Natural-Language Coding, and Agents
- OpenAI says Codex runs 'daily' across its engineering teams
- Meta's Muse Hits 500,000 Users in Week One, Faces Copying Claims