server
MCP server for Waiframe.ai β connect AI coding tools to your wireframe designs
Installation
npx @waiframe/mcp-serverAsk AI about server
Powered by Claude Β· Grounded in docs
I know everything about server. Ask me about installation, configuration, usage, or troubleshooting.
0/500
Reviews
Documentation
Waiframe MCP Server
Connect AI coding tools to your Waiframe.ai wireframe designs using the Model Context Protocol.
Build UI code that matches your wireframes β your AI coding tool gets structured context about screens, components, navigation flows, and design patterns.
Quick Start
1. Get an API Key
Go to waiframe.ai/app/settings?tab=api-keys and create an API key.
2. Configure Your AI Tool
Claude Code β add to your MCP settings:
{
"mcpServers": {
"waiframe": {
"command": "npx",
"args": ["-y", "@waiframe/mcp-server"],
"env": {
"WAIFRAME_API_KEY": "wf_sk_your_key_here"
}
}
}
}
Cursor β add the same config to your MCP settings file.
3. Use It
Once connected, your AI tool can:
- List your projects β see all wireframe projects
- Read screens β get component layouts in a semantic format (no UUIDs or pixel positions)
- Follow flows β understand navigation between screens
- Get design context β app type, audience, features, brand style
- Look up components β specs for any of the 60+ wireframe component types
Tools
| Tool | Description |
|---|---|
list_projects | List all your wireframe projects |
get_project_overview | Project details with screen and flow names |
get_screen | Semantic wireframe layout for a screen |
get_flow | Navigation flow between screens |
get_design_context | App context (type, audience, features) |
get_component_spec | Component specification and properties |
Resources
| Resource | Description |
|---|---|
waiframe://component-catalog | Full catalog of 60+ wireframe components |
waiframe://design-patterns | 20 curated UI design patterns |
How It Works
AI Coding Tool βstdioβ @waiframe/mcp-server βHTTPSβ waiframe.ai API
The MCP server runs locally on your machine. It connects to the Waiframe API using your API key and translates wireframe data into an AI-friendly format:
- Strips UUIDs, pixel positions, and sizes
- Keeps component types, properties, labels, and text
- Resolves navigation connections to screen names
- Sorts elements top-to-bottom for reading order
Example
A button in your wireframe becomes:
- button-primary {label: "Sign In"} β Dashboard
Instead of raw JSON with UUIDs and coordinates.
Environment Variables
| Variable | Required | Default | Description |
|---|---|---|---|
WAIFRAME_API_KEY | Yes | β | Your API key from waiframe.ai |
WAIFRAME_BASE_URL | No | https://waiframe.ai | API base URL (for development) |
Development
# Install dependencies
npm install
# Build
npm run build
# Run locally
WAIFRAME_API_KEY=wf_sk_... node dist/index.js
License
MIT
