KS
Killer-Skills

figma-plugin — Categories.community ai-enabled, figma-plugin

v1.0.0
GitHub

About this Skill

Perfect for Frontend Agents needing advanced UI design and prototyping capabilities with Figma plugins Config files for my GitHub profile.

# Core Topics

verygoodmurph verygoodmurph
[0]
[0]
Updated: 3/3/2026

Quality Score

Top 5%
34
Excellent
Based on code quality & docs
Installation
SYS Universal Install (Auto-Detect)
Cursor IDE Windsurf IDE VS Code IDE
> npx killer-skills add verygoodmurph/figma-plugin-starter/figma-plugin

Agent Capability Analysis

The figma-plugin MCP Server by verygoodmurph is an open-source Categories.community integration for Claude and other AI agents, enabling seamless task automation and capability expansion. Optimized for ai-enabled, figma-plugin.

Ideal Agent Persona

Perfect for Frontend Agents needing advanced UI design and prototyping capabilities with Figma plugins

Core Value

Empowers agents to develop and deploy custom Figma plugins using ES5 syntax, leveraging API recipes and theming capabilities outlined in AGENTS.md, while avoiding common pitfalls detailed in CLAUDE.md

Capabilities Granted for figma-plugin MCP Server

Developing custom Figma plugins for specific design tasks
Automating repetitive design workflows using Figma's API
Creating custom UI components and themes for Figma projects

! Prerequisites & Limits

  • Requires knowledge of ES5 syntax and Figma plugin development
  • Must adhere to two-context boundary rules
  • Limited to Figma plugin development only
Project
SKILL.md
4.4 KB
.cursorrules
1.2 KB
package.json
240 B
Ready
UTF-8
SKILL.md
Readonly

Figma Plugin Development

You are an expert Figma plugin developer. This starter template is a zero-build-step Figma plugin scaffold. Your job is to help users go from idea to working plugin.

Rules & Patterns

All rules (ES5 syntax, two-context boundary, pitfalls) are in CLAUDE.md (always loaded). All code patterns (API recipes, theming, manifest fields) are in AGENTS.md — read it before implementing.

Workflow

Follow the phase that matches what the user is asking for. If they have a new idea with no spec, start at Phase 1. If they already have a SPEC.md filled out, skip to Phase 2. If they want to add a feature to an existing plugin, go to Phase 3.


Phase 1: Spec Out the Plugin

Use this when the user has a plugin idea but no specification yet.

Goal: Produce a complete SPEC.md that's detailed enough to build from.

  1. Read the current SPEC.md, code.js, ui.html, and manifest.json to understand the starting point.

  2. Interview the user. Ask focused questions across these areas — skip anything already obvious from context:

    • Core purpose: What does the plugin do? What problem does it solve?
    • Target workflow: What's the user's step-by-step interaction with the plugin?
    • UI layout: What controls, displays, or panels does the plugin need? Approximate dimensions?
    • Figma API surface: Which Figma APIs are needed? (selection, variables, styles, text, nodes, etc.)
    • Data flow: What data moves between code.js and ui.html? What message types are needed?
    • External dependencies: Does it need network access? What domains?
    • Edge cases: What happens with empty selection, wrong node types, missing fonts, large documents?
    • Persistence: Does it need to store data? (clientStorage for user prefs, pluginData for per-node data)
  3. Ask questions iteratively — follow up on answers, don't dump everything at once. Stop when you have enough to build.

  4. Write the complete specification to SPEC.md with these sections:

markdown
1# [Plugin Name] — Specification 2 3## Overview 4One paragraph describing what the plugin does and why. 5 6## Features 7Numbered list of everything the plugin does. 8 9## UI Layout 10Description of the interface. Include dimensions (width x height). 11Describe each section, control, and interactive element. 12 13## Architecture 14 15### Message Passing Protocol 16Table of all message types: 17| Direction | Type | Data | Purpose | 18|-----------|------|------|---------| 19 20### Code Context (code.js) 21What the plugin thread does: event listeners, API calls, data processing. 22 23### UI Context (ui.html) 24What the UI does: rendering, user interaction, message sending. 25 26## Figma API Usage 27Which APIs are used and why. 28 29## Manifest Configuration 30Any changes needed to manifest.json (permissions, capabilities, domains). 31 32## Edge Cases & Error Handling 33How to handle each failure mode. 34 35## Accessibility 36Keyboard navigation, ARIA labels, focus management.

Phase 2: Build From Spec

Use this when SPEC.md is filled out and the user wants to build the plugin.

Goal: Transform the starter template into a working plugin matching the spec.

  1. Read SPEC.md thoroughly. Read AGENTS.md for API patterns and best practices.

  2. Plan the implementation: message types, API calls, UI components, build order.

  3. Update manifest.json — name, id, editorType, permissions, capabilities, allowedDomains.

  4. Implement code.js — showUI, event listeners, message handlers, API calls, error handling.

  5. Implement ui.html — HTML structure, Figma CSS tokens, interactivity, message passing, loading/error states.

  6. Validate:

    • Every message type in the spec has a sender AND receiver
    • Every Figma API call is wrapped in try-catch
    • Selection is checked before accessing selection[0]
    • Fonts are loaded before setting text
    • UI works in both light and dark theme
    • No ?. or ?? in code.js

Phase 3: Add a Feature

Use this when the user wants to add functionality to an existing plugin.

  1. Read the current code.js, ui.html, and manifest.json to understand what exists.

  2. Discuss the feature with the user: what it does, how it's triggered, what APIs it needs, any new message types.

  3. Create a SPEC-FEATURE-<NAME>.md file documenting the feature (kebab-case name).

  4. Implement: add message types to both contexts, new UI elements, new API calls with error handling, update manifest if needed.

  5. Verify the feature works without breaking existing functionality.

Related Skills

Looking for an alternative to figma-plugin or building a Categories.community AI Agent? Explore these related open-source MCP Servers.

View All

widget-generator

Logo of f
f

widget-generator is an open-source AI agent skill for creating widget plugins that are injected into prompt feeds on prompts.chat. It supports two rendering modes: standard prompt widgets using default PromptCard styling and custom render widgets built as full React components.

149.6k
0
Design

chat-sdk

Logo of lobehub
lobehub

chat-sdk is a unified TypeScript SDK for building chat bots across multiple platforms, providing a single interface for deploying bot logic.

73.0k
0
Communication

zustand

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication

data-fetching

Logo of lobehub
lobehub

The ultimate space for work and life — to find, build, and collaborate with agent teammates that grow with you. We are taking agent harness to the next level — enabling multi-agent collaboration, effortless agent team design, and introducing agents as the unit of work interaction.

72.8k
0
Communication