The Model Context Protocol (MCP) is an open standard that lets AI models interact with external tools and data sources. ToeBox uses MCP to connect your agents to Shopify and other services.Documentation Index
Fetch the complete documentation index at: https://docs.toebox.ai/docs/llms.txt
Use this file to discover all available pages before exploring further.
What is MCP?
MCP provides a standardized way for AI agents to:- Discover tools — Agents learn what tools are available and what they do
- Call tools — Agents can invoke tools with structured parameters
- Receive results — Tool outputs are returned in a format the agent understands
Transport: Streamable HTTP
ToeBox uses Streamable HTTP as the MCP transport protocol. This is simpler than WebSocket or SSE-based transports:- No persistent connections to manage
- Works through standard HTTP proxies and firewalls
- Each tool call is a single HTTP request/response
You do not need to configure transport settings. ToeBox handles the connection setup automatically.
Connecting Shopify MCP servers
ToeBox comes with pre-configured connections to Shopify’s official MCP servers:- Shopify Catalog
- Shopify Checkout
What it does: Product search and discovery across your Shopify store.Endpoint:
https://discover.shopifyapps.com/global/mcpAvailable tools:search_products— Search by keyword, filter by price, availabilityget_product_details— Retrieve full product informationbrowse_collections— List and browse product collections
Testing MCP connections
You can test MCP connections from the agent configuration page:- Open your agent’s settings
- Navigate to the MCP Tools section
- Each connected server shows a status indicator (connected/disconnected)
- Use the Test button to verify the connection is working
Exposing your agents via MCP
ToeBox also provides an MCP endpoint that lets external tools connect to your agents. This means you can use your ToeBox agents from:- Claude Desktop
- Claude Code
- Cursor
- VS Code
- Any MCP-compatible client