designer-skill
Click on "Deploy Server".
Wait a few minutes for the server to deploy. Once ready, it will show a "Started" state.
In the chat, type
@followed by the MCP server name and your instructions, e.g., "@designer-skillCreate a preflight brief for the new onboarding screen and gate the UI changes"
That's it! The server will respond to your query, and you can continue using it as needed.
Here is a step-by-step guide with screenshots.
Plug-and-play MCP. UI superpowers for your agent.
Setup ยท How it works ยท References ยท Tools ยท Development
Setup
Paste into your agent (Claude Code, Codex, Cursor, any MCP client). It installs the skill, wires the MCP and verifies the tools:
Add designer-skill alongside Niblet Designer UI. It is optional and keyless: it routes a design request, loads the right reference, and gates the ship. Niblet supplies the real screens and materials.
1. Install or refresh the skill into the user skill directory (not this project):
npx skills add PyModel/designer-skill --skill designer-skill -g -y
2. Add the local MCP using this client's native MCP configuration, in the user-level (not project-level) config:
Name: designer-skill
Transport: stdio
Command: npx
Args: -y @pymodel/designer-skill-mcp@latest
No API key is needed. Do not add one.
3. Verify: list the designer-skill server's tools and confirm get_preflight_brief, dispatch_intent, review_and_gate are present. Do not run a sample check.
When setup is complete, confirm that the designer-skill skill is installed and that the three tools are exposed. On UI work from now on: call get_preflight_brief first, use dispatch_intent to load only the references it recommends, pull screen references and materials from the niblet server, then call review_and_gate on the changed files before saying the work is done. If there is an active UI task, continue it. Otherwise say: "designer-skill is ready and paired with Niblet."No Niblet? Drop the "pull screen references and materials from the niblet server" clause.
Or pick one:
Install | |
๐ฃ Claude Code plugin (skill + MCP) |
|
๐ข Codex plugin |
|
โซ Cursor plugin | Install from the marketplace; ships |
๐ต MCP only, any client |
|
๐ Skill only |
|
Canonical MCP config (mcp.json):
{ "mcpServers": { "designer-skill": { "command": "npx", "args": ["-y", "@pymodel/designer-skill-mcp@latest"] } } }@latest tracks npm; teams pin @pymodel/designer-skill-mcp@0.21.1. Plugin skill content updates separately (/plugin update โฆ). MCP registry name: io.github.PyModel/designer-skill-mcp. Requires Node 22+.
VS Code .vscode/mcp.json (1.99+):
{ "servers": { "designer-skill": { "type": "stdio", "command": "npx", "args": ["-y", "@pymodel/designer-skill-mcp"] } } }Codex CLI ~/.codex/config.toml:
[mcp_servers.designer-skill]
command = "npx"
args = ["-y", "@pymodel/designer-skill-mcp"]Open Code opencode.json:
{ "mcp": { "designer-skill": { "type": "local", "command": ["npx", "-y", "@pymodel/designer-skill-mcp"] } } }Claude Desktop, Cursor (.cursor/mcp.json), Kilo Code (mcp_settings.json), Pythinker (~/.pythinker/mcp.json): the canonical mcpServers JSON above. Pythinker verify: pythinker mcp test designer-skill (guide).
Pi: the same JSON, or register the skill natively: { "skills": [{ "path": "/path/to/skills/designer-skill/SKILL.md" }] }
Local checkout: replace npx with "command": "node", "args": ["/abs/path/to/designer-skill-mcp/dist/index.js"].
Related MCP server: CodeCrafted Design MCP
How it works
flowchart LR
A[Prompt] --> B[get_preflight_brief]
B --> C[dispatch_intent]
C --> D[get_reference]
D --> E[Build / polish UI]
E --> F[review_and_gate]
F -->|FAIL: fix| E
F -->|static checks pass| G[Rendered checks<br/>by the host]
classDef start fill:#f8fafc,stroke:#94a3b8,color:#18181b
classDef route fill:#eff6ff,stroke:#3b82f6,color:#1e3a8a
classDef know fill:#faf5ff,stroke:#7c3aed,color:#4c1d95
classDef work fill:#fff7ed,stroke:#e87a3d,color:#7c2d12
classDef gate fill:#ecfdf5,stroke:#10b981,color:#065f46
classDef host fill:#fff1f2,stroke:#e11d48,color:#881337
class A start
class B,C route
class D know
class E work
class F gate
class G host๐ต Route | ๐ฃ Know | ๐ข Check |
| 16 designer references (type, color, motion, a11y, anti-slop, redesign) plus 26 | A 44-rule deterministic detector backs |
The gate is static only: overall status is FAIL or NOT_VERIFIED, never a rendered-readiness pass. Rendered, accessibility and performance checks stay NOT_RUN until the host supplies evidence.
Example: "Use designer-skill to redesign this pricing page without breaking functionality." โ get_preflight_brief โ dispatch_intent โ get_reference โ edit โ review_and_gate.
Built with it: pythinker.com
References
File | Use when | Tier |
| Typography, spacing, color, layout, hierarchy | ๐ต core |
| Being distinctive: inverse test, layout menu, named references | ๐ต core |
| Picking a look: 5 systems with palettes, fonts, shadows | ๐ต core |
| Timing, springs, scroll, reduced motion | ๐ต core |
| Tokens, a11y, responsive, Core Web Vitals | ๐ต core |
| Ban list, category-reflex checks, completeness contract | ๐ต core |
| Audit โ diagnose โ redesign without breaking behavior | ๐ต core |
| Intent โ verb dispatch | ๐ต core |
| Evidence rules, gate statuses, failure triage | ๐ต core |
| Fitts/Hick/Miller, forms, navigation, errors, loading | ๐ด extended |
| Seven-dimension critique | ๐ด extended |
| Token architecture, component specs, theming | ๐ด extended |
| Discovery interview, PRODUCT.md, DESIGN.md | ๐ด extended |
| Shape-then-build pipeline with user gates | ๐ด extended |
| Browser variant mode: select, HMR, steer, accept | ๐ด extended |
| Modern CSS: container queries, | ๐ด extended |
Plus 26 ux/* references in skills/ux-designer/. Each SKILL.md is a short router with a one-line description; references load only when a task needs them.
Phrase | Verbs | Reads |
"make it pop" |
|
|
"it feels off" |
|
|
"production-ready" |
|
|
"add some motion" |
|
|
"it looks AI-made" |
|
|
"redesign this" |
|
|
Tools
Tool | Purpose |
| Scope and verification contract (call first) |
| Read PRODUCT.md / DESIGN.md from the project (absolute |
| SKILL.md router and reference map |
| Load one reference by name (designer or |
| Advisory style and truthful-content review guidance |
| All design verbs with descriptions |
| Help and reference names for one verb |
| Map a request โ verb(s) + at most four references to read |
| Validate a context-grounded direction record |
| OKLCH brand seed for authorized new palette work |
| Deterministic static scan (44 rules): coverage, file hashes, gaps |
| Static gate per required rule; never claims rendered readiness |
| Optional niblet real-screen search ( |
| Optional niblet structured reference ( |
Resources: designer://skill ยท designer://reference/{+name} ยท Prompt: design (task, optional aesthetic)
Development
cd designer-skill-mcp
npm ci
npm run build # syncs skills/ โ assets/ (generated, gitignored), compiles TypeScript
npm run typecheck && npm test # tsc over src + test, then vitest; `npm run smoke` drives the packed tarballHTTP mode (Streamable HTTP at /mcp):
node dist/index.js --http --port 3017 --root /abs/project # 127.0.0.1
DESIGNER_SKILL_HTTP_TOKEN=โฆ node dist/index.js --http --host 0.0.0.0 --root /abs/projectEvery HTTP bind requires at least one --root. Loopback binds validate the Host header; a non-loopback bind also requires DESIGNER_SKILL_HTTP_TOKEN (Authorization: Bearer โฆ).
Release: ./scripts/release.sh "notes" bumps and syncs every version, verifies, tags and pushes; publish.yml publishes npm (with provenance), the MCP registry entry and the GitHub release. Contract details: docs/HARDENING.md.
This server cannot be deployed
Maintenance
Related MCP Connectors
Design intelligence for coding agents: audits, design systems, and a taste profile agents consult.
Live React design-system APIs, patterns, and code validation so AI agents build real UI, not slop.
UI design from prompts, screenshots, and URLs for AI coding agents and theme tokens.
Turn PRDs and product ideas into structured specs so coding agents build your intent, not theirs.
Related MCP Servers
- AlicenseNot gradedqualityAmaintenanceSTOP UI SLOP. Gives coding agents searchable evidence from 800,000+ real web and iOS screens, design contracts, and a hard UI finish gate.1MIT
- FlicenseNot gradedqualityCmaintenanceEnables AI coding agents to plan, build, and review websites and product interfaces with a persistent, user-led process, including design direction, component contracts, and implementation review.-
- AlicenseNot gradedqualityBmaintenanceProvides design systems, UI prompts, and layout variation guidance to AI coding tools for generating better user interfaces.233 npm1,955MIT
- AlicenseNot gradedqualityAmaintenanceEnables AI agents to create and manage consistent multi-screen UI designs through a token-efficient MCP interface, with design system tokens, components, flows, and visual review.336 npm1AGPL 3.0