A Model Context Protocol (MCP) server for Analyzify — the Shopify analytics and tracking app (Shopify App Store).
Connects AI assistants to your Shopify store data, Google Analytics 4, Google Search Console, Google Ads, and historical analytics reports through natural conversation.
curl -sL https://raw.githubusercontent.com/analyzify/analyzify-mcp-public/main/install.sh | shirm https://raw.githubusercontent.com/analyzify/analyzify-mcp-public/main/install.ps1 | iexDownload from the Releases page, extract, and place in your PATH.
analyzify-mcp versionThe easiest way to configure Claude Desktop is with the setup command:
analyzify-mcp setupThis will ask for your API key and automatically update your Claude Desktop config. You can also pass the key directly:
analyzify-mcp setup --api-key azfy_your-api-keyFor Claude Code:
analyzify-mcp setup --claude-code --api-key azfy_your-api-keyAfter setup, restart Claude Desktop to activate.
The Analyzify skill teaches your AI assistant how to use the MCP server effectively — what queries to run, how to present data, and how to handle different analytics questions.
- Download the latest
analyzify-skill-v*.zipfrom Releases - Open Claude Desktop and go to Settings > Customize > Skills
- Click + then Create skill then Upload a skill
- Upload the zip file (no need to extract it)
See Claude's skill documentation for more details.
- Download the latest
analyzify-skill-v*.zipfrom Releases - Extract the zip
- Upload the skill files to your project's custom instructions or knowledge
npx skills add analyzify/analyzify-mcp-publicIf you prefer to configure manually instead of using analyzify-mcp setup:
Claude Desktop (claude_desktop_config.json):
{
"mcpServers": {
"analyzify": {
"command": "analyzify-mcp",
"args": ["serve"],
"env": {
"ANALYZIFY_API_KEY": "azfy_your-api-key"
}
}
}
}Cursor / VS Code (MCP settings):
{
"analyzify": {
"command": "analyzify-mcp",
"args": ["serve"],
"env": {
"ANALYZIFY_API_KEY": "azfy_your-api-key"
}
}
}Without the API key, the server works in schema-only mode.
Profit & Revenue: "What's my profit?", "MER?", "P&L breakdown?"
Diagnostics: "Why are my sales down?", "What changed since last month?"
Reports: "Give me a weekly report", "Monthly business summary"
Ads & ROAS: "How are my Google Ads performing?", "Best/worst campaigns?"
Attribution: "Where do my orders come from?", "Channel attribution?"
Products: "Best selling products?", "Which products should I promote?"
Traffic & SEO: "Top search queries?", "Organic vs paid traffic?"
Conversion: "How do I improve my conversion rate?", "Cart abandonment?"
Customers: "New vs returning customers?", "Customer acquisition cost?"
Seasonal: "Compare to last year?", "Am I on track for my annual target?"
Promotions: "Did my sale help?", "Are discounts cutting into margin?"
Tracking: "Is my conversion tracking working?", "Last Google Ads conversion?"
analyzify-mcp setup # Configure Claude Desktop automatically
analyzify-mcp serve # Start MCP server (stdio)
analyzify-mcp serve --transport sse --addr :8080 # Start MCP server (SSE HTTP)
analyzify-mcp version # Print version
analyzify-mcp self-update # Update to latest version
analyzify-mcp self-update --check # Check for updates
analyzify-mcp skill latest-version # Check latest skill version
analyzify-mcp search proxy # Search schema
analyzify-mcp query '{ extMe { shopName } }' # Execute a query| Tool | Description |
|---|---|
introspect_schema |
Search the schema for types, queries, and mutations |
graphql_schema_full |
Returns the complete schema in SDL format |
execute_graphql |
Execute a GraphQL query against the live API |
introspect_report_schema |
Search the report schema for available report queries |
report_schema_full |
Returns the complete report schema in SDL format |
execute_report_graphql |
Execute a query against the Report API |
macOS / Linux:
sudo rm /usr/local/bin/analyzify-mcpWindows (PowerShell):
Remove-Item "$env:LOCALAPPDATA\analyzify-mcp" -Recurse -Force