MCP Server
Connect your AI assistant to Kleio for conversational store analytics
What is MCP?
MCP (Model Context Protocol) is an open standard that lets AI assistants connect to external tools and data sources. Kleio’s MCP server gives your AI assistant direct access to your store’s P&L metrics — so you can ask questions like “What was my profit last month?” and get an instant answer.
Available Tools
Once connected, your AI assistant gets access to the following tools:
- list_stores — See all your connected Shopify stores with their currency and timezone settings
- get_metrics — Dashboard and P&L metrics like revenue, CM1/CM2/CM3, ad spend, etc.
- get_product_metrics — Product-level data grouped by product, variant, vendor, collection, and more
- get_product_ltv_metrics — Product-level lifetime value analysis showing how first-order products drive repeat margin over 12 months
- get_cohort_metrics — Monthly customer acquisition cohort analysis with repeat rates and LTV progression for both margin and revenue
- get_country_metrics — Country-level P&L breakdown for comparing profitability across markets
- get_inventory — Current inventory status with stockout forecasting, reorder signals, and sell-through rates
- get_timings — Time-between-orders analysis showing repurchase cycles by product
All tools support filters by date range, sales channel (online/POS), customer type (new/existing/subscription), country, and order tags.
Analytics Playbook
Every time your AI assistant connects to Kleio’s MCP server, it automatically receives our analytics playbook — a structured guide that teaches it how to think about your store the way an experienced operator would.
Without the playbook, your AI has data but no judgment. With it, the AI knows which metrics actually answer the question you’re asking, when to combine multiple data sources to get the full picture, and when a surface-level number is hiding something worth digging into.
Connecting
Kleio’s MCP server URL:
https://app.getkleio.com/api/mcp
Authentication is handled automatically via OAuth — you’ll be prompted to log in to your Kleio account the first time you connect.
Claude Desktop
- Open Settings → Connector → Add custom connector
- Insert the MCP server URL:
https://app.getkleio.com/api/mcp - You’ll be prompted to authorize Kleio on first use
Some users may need to restart Claude Desktop after connecting for the connection to take effect.
Claude Code
- Run the following command:
claude mcp add kleio --transport http https://app.getkleio.com/api/mcp
- You’ll be prompted to authorize Kleio on first use
Other MCP Clients
Any MCP-compatible client can connect to Kleio. Point it to the server URL (https://app.getkleio.com/api/mcp) using the HTTP transport. The OAuth consent flow will handle authentication automatically.
High-Value Prompts to Ask Right Away
These prompts show the kind of deep analysis you get immediately after connecting:
- “What are our best products for acquiring new customers?”
- “Do we have any dead stock I need to liquidate?”
- “What products should we push in the near future?”
- “When is the optimal time for us to re-engage customers?”
- “Analyze our P&L and recommend how we can make more money.”