ServiceNow Uses Figma Files as Prompts to Generate Enterprise Apps

They say that a picture is worth a thousand words, and ServiceNow is putting that to the test when it comes to building the next generation of enterprise applications. The company has announced that it now supports Figma in its Build Agent, allowing Figma design files to be used as prompts for building apps, rather than relying solely on text-based instructions. In doing so, ServiceNow bridges the persistent gap between design and development.

“Great product design makes a positive human impact when it becomes a real experience in the hands of people,” Amy Lokey, ServiceNow’s chief experience officer and executive vice president, remarked in a statement. “By connecting Figma with ServiceNow’s AI-powered workflows, we’re making it effortless for teams to turn ideas into enterprise-ready applications in minutes, so they can focus on what matters most: creative human-centered experiences with enterprise-scale and impact.”

Introduced in September, ServiceNow’s Build Agent is the company’s answer to Cursor, GitHub Copilot, Vercel, Lovable, Replit, and Windsurf. It’s part of a trend by enterprise software vendors to provide vibe-coding tools to customers—Salesforce has its recently released Agentforce Vibes, and SAP has SAP Build. All of these solutions are primarily developer-centric, but ServiceNow appears to be considering the broader picture—how good design contributes to practical applications that teams will want to use.

Incorporating image files—even those with multiple layers—can provide AI tools, such as ServiceNow’s Build Agent, with the necessary context to follow through on the prompt instructions and avoid hallucinations. No longer will users be subjected to the “What I Expected/What I Got” meme of development. Now, the bot is inching closer to being like a mind reader and bringing ideas into reality.

It’s one thing to ask AI to generate an app’s functionality, but the user experience is a different challenge. How do you describe a layout or interaction model well enough for a bot to design it accurately? Some nuances are best conveyed visually, and existing vibe coding tools aren’t equipped to deal with them today. ServiceNow is partnering with Figma to help prevent any loss of translation when it comes to the AI experience, especially as it fosters human-AI collaboration.

Thanks to its connection to Figma’s Model Context Protocol (MCP) server, ServiceNow’s Build Agent can interpret a design’s layout and components to generate a fully functional application automatically. This integration is governed by both ServiceNow and Figma’s security framework, using OAuth 2.0 authentication and secure, server-to-server communication. All applications will natively have ServiceNow’s permissioning, audit trails, and version controls.

“In a world of AI-generated software, design is the differentiator that will make your product stand out,” Kris Rasmussen, Figma’s chief technology officer, said in a release. “This MCP integration brings important Figma design context directly into ServiceNow’s AI workflows to help teams efficiently build high-quality, differentiated enterprise products.

Now that ServiceNow has announced its Figma collaboration, it’s only a matter of time before other software vendors make similar moves. It may not only be with Figma, either—it could be with Adobe (whenever it launches an MCP server) or Canva.

In the meantime, anyone wishing to take advantage of this capability can do so with the latest Build Agent release in the ServiceNow Store.

Featured Image: ServiceNow signage at Knowledge 2025. Credit: Ken Yeung