axure-mcp-server
Extracts text, image links, and structural content from public Axure share pages to enable AI-powered prototype summarization. It features multi-page crawling and OCR fallback to process image-only designs for downstream LLM analysis.
README
axure-mcp-server
MCP server for extracting visible text and image links from public Axure share pages, so AI tools (Cursor/Claude Desktop/others with MCP support) can summarize prototype content.
Features
- Extract visible text blocks from Axure page DOM
- Extract image URLs (optionally image base64 payload)
- Auto-discover and crawl additional same-origin Axure pages
- OCR fallback when text is image-only or too little
- Return structured JSON for downstream AI summarization
- Build an AI-ready summary prompt in one call
- Works with MCP clients via stdio
Requirements
- Node.js >= 18
- npm >= 9
If your local Node is old (for example Node 16), upgrade first:
# if you use nvm
nvm install 20
nvm use 20
node -v
Quick Start
npm install
npx playwright install chromium
npm run build
npm start
MCP Tools
1) axure_health
Health check of runtime.
2) axure_fetch
Input:
{
"url": "https://vscn2w.axshare.com/?id=xpnh6e&p=%E5%8E%9F%E5%9E%8B%E6%96%B9%E6%A1%88&sc=3",
"timeoutMs": 45000,
"maxImages": 30,
"maxTexts": 300,
"crawlPages": true,
"maxPages": 5,
"enableOcrFallback": true,
"ocrMinTextCount": 8,
"ocrMaxImages": 3,
"ocrLanguage": "chi_sim+eng",
"includeImageBase64": false
}
Output: JSON with status, textBlocks, imageItems, pages, warnings, stats (including ocrTextCount).
3) axure_summary_prompt
Input:
{
"url": "https://vscn2w.axshare.com/?id=xpnh6e&p=%E5%8E%9F%E5%9E%8B%E6%96%B9%E6%A1%88&sc=3",
"focus": "请提炼核心流程和页面功能点",
"crawlPages": true,
"maxPages": 6,
"enableOcrFallback": true
}
Output: an AI-ready plain text prompt including extracted texts and image links.
Use in Cursor
Add to MCP config (example):
{
"mcpServers": {
"axure-mcp": {
"command": "node",
"args": ["/ABSOLUTE/PATH/axure-mcp-server/dist/index.js"]
}
}
}
If published to npm:
{
"mcpServers": {
"axure-mcp": {
"command": "npx",
"args": ["-y", "axure-mcp-server"]
}
}
}
Compatibility: keep Node 16 globally, run MCP on Node 20
If your main frontend stack (for example Vue2) must stay on Node 16, you can still run this MCP safely by pinning only this server to Node 20 in Cursor:
{
"mcpServers": {
"axure-mcp": {
"command": "npx",
"args": [
"-y",
"node@20",
"/Users/55haitao/Desktop/axure-mcp-server/dist/index.js"
]
}
}
}
This keeps your global Node unchanged while ensuring axure-mcp-server runs with a compatible runtime.
Compatibility after npm publish
If you publish this package to npm, and still need to keep global Node 16, use Node 20 only for this MCP process:
{
"mcpServers": {
"axure-mcp": {
"command": "npx",
"args": [
"-y",
"node@20",
"/usr/local/bin/npx",
"-y",
"axure-mcp-server"
]
}
}
}
If your system npx path is different, replace "/usr/local/bin/npx" with your actual path from which npx.
Publish to GitHub
git init
git add .
git commit -m "feat: init axure mcp server"
git branch -M main
git remote add origin <your-repo-url>
git push -u origin main
Publish to npm
Before publish:
- Update
package.jsonfields (name,author,repository,homepage) - Ensure build output exists:
npm run build
Then:
npm login
npm publish --access public
Roadmap
- Add optional multi-page navigation and auto-click flow
- Export markdown report and downloadable screenshot package
Notes
- Designed for public/no-login Axure links.
- Respect source site terms and data usage permissions.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.