No red flags found in any of the 11 categories — no credential harvesting, no data exfiltration, no curl-pipe-shell installer. Scanned against the SlowMist agent-security taxonomy, refreshed every 8 hours. Full audit →
by natdexterra · MCP Server · ★ 51
Last updated: · Indexed by AgentSkillsHub · Auto-synced every 8h
🔒 Is work-with-design-systems safe to install? View the security audit →
work-with-design-systems A Claude skill for working with Figma design systems — two modes in one place: inspect (read-only audits with WCAG checks, component scoring, handoff docs) and build (creating components, fixing foundations, adding slots, writing descriptions). Build mode optionally extends to Phase 6 (sync to code) — generates , an audit script, and AI rules file for your codebase. Works with Claude Code, Cursor, and Codex. What's new in v2.0 This is a major release. The skill was previously called and only handled building. Now it covers the full lifecycle: inspect existing files, decide what to build or fix, optionally export to code. Between inspect and build, the skill always pauses for your decision. Phase 6 is OFF by default and only triggers on explicit request. Migration from v1.x: see CHANGELOG for the migration note. Repo URL has changed but old links redirect automatically. Highlights: Two modes — inspect (read-only) and build (write), with mandatory pause between Phase 6 — sync to code (optional, off by default) — generates with three-layer indirection, AI rules file (Claude Code / Cursor / Codex), and CI-ready audit script. Closes the design-to-code loop.
| Stars | 51 |
| Forks | 4 |
| Language | JavaScript |
| Category | MCP Server |
| License | MIT |
| Quality Score | 68.3175161950557/100 |
| Last Updated | 2026-09-17 |
| Created | 2026-04-01 |
| Platforms | claude-code, mcp, node |
| Est. Tokens | ~18k |
These tools work well together with work-with-design-systems for enhanced workflows:
Looking for a work-with-design-systems alternative? If you're comparing work-with-design-systems with other mcp server tools, these 6 projects are the closest alternatives on Agent Skills Hub — ranked by topic overlap, star count, and community traction.
Local Figma MCP server with no REST rate limits, no metered tool-call quotas, and a verification loop. Drop-in
Design in Figma with Claude Code. Bridge connects your terminal to the Figma Plugin API via WebSocket.
The most powerful Figma Console MCP capabilities, repackaged as Markdown skills for the native Figma MCP serve
AI can draw UI directly on the Figma canvas via JavaScript, and read existing designs back as structured data.
Extract design tokens (colors, typography, spacing, border radius, shadows) from any public website. Generates
Claude Code skills for the work that keeps a design system alive
Explore other popular mcp server tools:
work-with-design-systems is Inspect, build, or extend Figma design systems from Claude Code. Read-only audits with WCAG checks and component scoring, build mode with variable bindings and slots, optional Phase 6 sync to tokens.c. It is categorized as a MCP Server with 51 GitHub stars.
work-with-design-systems is primarily written in JavaScript. It covers topics such as agent-skills, claude-code, design-system.
You can find installation instructions and usage details in the work-with-design-systems GitHub repository at github.com/natdexterra/work-with-design-systems. The project has 51 stars and 4 forks, indicating an active community.
work-with-design-systems is released under the MIT license, making it free to use and modify according to the license terms.
The top alternatives to work-with-design-systems on Agent Skills Hub include plumb-mcp, bridge, figma-console-mcp-skills. Each offers a different approach to the same problem space — compare them side-by-side by stars, quality score, and community activity.
Grades come from a rule-based scan built on the SlowMist agent-security taxonomy, covering 11 red-flag categories including credential harvesting, data exfiltration, and curl | sh installers. It is a first-layer scan, not a manual audit — we say so rather than overstate it.
The scale of the problem is documented independently: Liu et al. (2026), in a study of 31,132 agent skills, report that 26.1% contain security vulnerabilities. Our own full-catalog census is published as a citable open dataset.
Sources & who's responsible: