Figma Make: Custom Skills for Reusable Prompt Workflows
FigmaFigma introduced Custom Skills in Figma Make, enabling designers and teams to create reusable markdown-based workflows invoked via slash commands directly inside Make's chat interface. A skill encodes a team's conventions and repeatable steps — for example, /insert-sample-data pulls in company-approved test data, while /build-from-prd paired with a Notion or Confluence connector converts a PRD into a standards-compliant prototype. Skills are currently personal and managed individually, with team and org-level sharing planned for the near future.
Featured Video
A video we selected to help illustrate this changelog
Sources & Mentions
5 external resources covering this update
Figma Make Gets Custom Skills: Reusable Workflow Templates via Slash Commands
Figma expanded its AI-powered prototyping tool Make on May 11, 2026 with the introduction of Custom Skills — a way for designers to package their own repeatable workflows into reusable, slash-command-accessible routines. Rather than re-explaining context and conventions in every new prompt session, teams can now define a skill once and invoke it consistently across all their Make files.
What Are Custom Skills?
Custom Skills in Make are markdown files that encode a designer's or team's standards, conventions, and step-by-step workflows. Figma's concept of "skills" originated in the MCP server ecosystem (where skills teach AI agents how to interact with the design canvas), but Custom Skills bring this idea directly into the Make chat interface — no developer setup required.
Each skill acts as a compressed prompt package. When invoked, it gives Make's AI a stable, ordered set of instructions to follow, reducing drift and improving prototype consistency across sessions. Rather than producing vague "AI-flavored" output, Make follows the team's actual design standards.
How They Work
Users can either import existing skills from the Figma Community or create their own directly within Make. Once a skill is created, it is accessible in any Make file via a slash command — for example, typing /insert-sample-data drops in company-approved test data on demand, and /build-from-prd (when paired with a Notion or Confluence connector) automatically turns any product requirements document into a functioning prototype that matches team standards.
The underlying markdown structure for a skill typically includes a clear title and purpose describing when the skill should be used, numbered instructions written in imperative language, representative examples of expected inputs and outputs, and edge case handling so the AI knows how to recover when the ideal path fails.
Current Availability and Future Plans
As of the May 11 release, each person creates and manages their own skills independently. Figma has signaled that team and organization-level skill sharing — allowing a design system owner or lead designer to publish and distribute skills across their org — is coming soon. A curated collection of community-built skills is available to browse on the Figma Community Skills page.
Why It Matters
Custom Skills represent a meaningful shift in how teams can standardize AI-assisted prototyping. Previously, each Make session started from scratch — designers had to re-explain component conventions, naming patterns, content guidelines, and workflow preferences in every prompt. Custom Skills eliminate that overhead, acting as institutional memory that the AI can reliably follow.
The feature is especially powerful when combined with Make's existing connectors (Notion, Confluence, Zapier, and others), since a skill can orchestrate multi-step workflows that pull in external context and apply it within Figma's design environment automatically.