Scan e-commerce stores for AI shopping readiness from Claude Code, Claude Desktop, Cursor, and any MCP-compatible AI environment.
Install in seconds
claude mcp add ucpready -- npx -y ucpready-mcpSetup
Terminal
claude mcp add ucpready -- npx -y ucpready-mcpclaude_desktop_config.json
{
"mcpServers": {
"ucpready": {
"command": "npx",
"args": ["-y", "ucpready-mcp"]
}
}
}.cursor/mcp.json
{
"mcpServers": {
"ucpready": {
"command": "npx",
"args": ["-y", "ucpready-mcp"]
}
}
}Capabilities
scan_storeRun a complete 4-pillar scan: Protocol Support, Crawler Access, Data Quality, and AI Visibility. Returns score, grade, and prioritized issues with fixes. Takes 15-30 seconds.
check_schemaCheck Product markup on a single page. Returns a pass/fail checklist of fields: name, description, price, images, GTIN, brand, and reviews.
check_crawlersTest which AI crawlers (GPTBot, ClaudeBot, PerplexityBot, etc.) are allowed or blocked by a site's robots.txt.
generate_llms_txtGenerate an llms.txt file for any website. Crawls pages and creates a structured discovery file for AI models.
Architecture
The MCP server connects your AI assistant to UCPReady.ai's scanning API. When you invoke a tool, the server sends requests to ucpready.ai, processes the results, and returns a formatted report directly in your conversation. No API keys required — the same free tier limits apply as the web tools.
FAQ
MCP is an open protocol that connects AI assistants like Claude to external tools and data sources. It allows AI coding environments to call specialized tools — like UCPReady's scanner — directly from your conversation, without leaving your workflow.
No. The MCP server uses UCPReady.ai's free public API. The same rate limits apply as the web tools — 3 full scans per day and 15 tool checks per day per IP address. No signup or authentication required.
Any MCP-compatible client works, including Claude Code (Anthropic's CLI), Claude Desktop, Cursor, Windsurf, and other editors that support the Model Context Protocol. Installation takes one command.
Any public e-commerce store URL. The full scan checks all 4 pillars of AI readiness: Protocol Support (UCP manifest), Crawler Access (robots.txt), Data Quality (schema.org markup), and AI Visibility. You can also run individual checks for schema validation, crawler access, or llms.txt generation.
Yes. The ucpready-mcp package is free to install and use. It's published on npm and the source code is available in the UCPReady GitHub repository.
Don't use AI coding tools? Scan any store directly on our website — free, instant results.
Scan Your Store Free