
Figma announced on LinkedIn an update to its Model Context Protocol server, expanding what developers and AI tools can automate inside the design platform. The update introduces the ability to generate and edit presentation decks in Figma Slides, build data-backed boards in FigJam, and make roundtrip edits between code and canvas in Figma Make. The company also added support for custom fonts and improved image exports, addressing key quality gaps in automated design workflows.
MCP Server Brings AI Workflows Into Figma
The Model Context Protocol, developed by Anthropic, allows AI assistants like Claude to connect directly to external tools and data sources. Figma's MCP server implementation means developers can now script or AI-generate design work across Figma's product suite without manually opening files or clicking through the interface.
The latest update covers three core use cases. First, teams can automate deck creation in Figma Slides, useful for generating recurring reports or structured presentations from templates and data. Second, FigJam boards can now be populated programmatically with structured data, turning spreadsheets or API responses into visual boards. Third, developers working in Figma Make can now push code changes back to the canvas and pull design updates into their codebase, creating a more fluid handoff between design and development.
Figma Make recently gained codebase integration and version control, and this MCP update tightens that loop further by allowing automated syncing via external scripts or AI agents.
Custom Fonts and Image Exports Close Quality Gaps
Two smaller but critical additions round out the update. Custom font support means AI-generated designs no longer default to system fonts, preserving brand consistency when automating layouts. Previously, automated workflows often broke typography rules or required manual cleanup. Now, MCP-driven designs can respect a team's typeface library from the start.
Image export improvements let developers pull high-quality assets directly from Figma files via the MCP interface. This matters for teams building marketing automation, documentation pipelines, or any workflow that needs to programmatically extract visuals at scale. Instead of exporting manually or using Figma's REST API separately, the MCP server handles it in the same context as other design operations.
Together, these updates make Figma more viable as a backend for automated design systems, not just a canvas for human designers.
Why It Matters
Figma's MCP server positions the platform as infrastructure for AI-assisted design workflows, not just a tool teams use manually. Companies building internal design automation, marketing agencies generating client decks at scale, or product teams syncing design and code repos can now script these tasks end-to-end.
The timing aligns with Figma's broader push into code and automation. Following the launch of its embedded Design Agent and local code editing support in Make, the MCP update extends Figma's reach beyond manual workflows. For small teams and solo builders, this means fewer repetitive tasks. For enterprises, it means design operations can scale without hiring more people to copy-paste data into slides or sync specs between tools.
The addition of custom fonts and image exports suggests Figma is serious about closing the quality gap between human-made and machine-generated designs. Automated workflows often fail on details like typography and asset resolution. By addressing these upfront, Figma makes it easier to trust AI-generated output without extensive review.
Developers interested in the Figma MCP server can access documentation and examples at Figma's developer hub. The update is live now for all users running MCP-compatible tools.
Sources
1 checkedHow we cover tool news: Create With's tool desk drafts these reports with AI from the sources listed above and checks them against those sources before publishing.







