Self-hosted alternatives to openai/figma-use

O

openai/figma-use

Prerequisite skill for every usefigma tool call

figma-use is a mandatory prerequisite skill from OpenAI's curated skills collection. It must be loaded before every call to the use_figma MCP tool, because that tool runs JavaScript inside a Figma file, and the rules it needs are not obvious. The skill loads the Plugin API rules, the critical gotchas and the reference documents that must be in context before any code runs. Those gotchas cover the handling of return values, the page context, reassigning fills, and loading fonts before editing text, and the skill notes that calls fail in ways that are hard to diagnose when these rules are missed. It covers node creation, variable binding, auto-layout, components and multi-step workflows. Typical uses are loading the Plugin API rules before creating Figma components from code, running JavaScript to set up design tokens and variable collections, building multi-section layouts with auto-layout and component instances, binding design system variables to properties such as fills and spacing, and inspecting a file's structure before any canvas change is made. It is not a design tool by itself. It is the reference an assistant reads so that the other Figma skills work reliably. Because it is a prerequisite, it adds no output of its own, and it is most useful when used with the figma-generate-design or figma-generate-library skills. The skill is listed on officialskills.sh and installs from the openai/skills repository. Install it with npx skills add https://github.com/openai/skills --skill figma-use.