What is MCP?
The Model Context Protocol (MCP) is an open standard that lets an AI assistant (Claude, ChatGPT, Gemini, Copilot...) connect to an external service and use its data as a tool, in natural language, right in the middle of a conversation, no development work or manual export required.
spyneMCP (https://mcp.spyne.app/mcp) exposes your data to the AI assistant of your choice: campaigns, keywords, site analytics, goals, custom dimensions, workspaces... The assistant queries spyne on your behalf and answers directly in your conversation.
What you can do with it
The MCP gives you access to your raw data in your workspaces. Specifically:
Query raw data : ask for precise data analysis (e.g. media performances by campaign x country x week).
Extract your data in automated format: a table to paste into a report, a summary text, a CSV, a powerpoint template... handy for automating a recurring task without starting from a manual export every time.
Ask quick questions about your campaigns, without opening spyne: "What is my spend last week on META campaigns ?", "Is my ROAS goal on track ?"
Manage and diagnose your configuration (custom dimensions, analytics matching rules, workspaces) in natural language. Coming soon.
The MCP queries raw data, not your existing Spyne reports.
How to connect it
spyneMCP connects in a few minutes to most AI assistants on the market: Claude, ChatGPT, Gemini, Dust, Microsoft Copilot.
Server URL:
https://mcp.spyne.app/mcpStep-by-step guide per platform : How to Connect the Spyne MCP to Claude, ChatGPT, Gemini, Dust, and Microsoft Copilot
Go further
Once connected, the best way to discover the MCP is to try real prompts. See the companion "spyneMCP : Use Cases" article for ready-to-copy examples, organized by use case.