Google Analytics MCP Server

The Google Analytics MCP server provides tools for LLMs to interact with Google Analytics via the Admin and Data APIs. The available tools are categorized into three functional groups:

Account and Property Information

• get_account_summaries: Retrieves information about the user's Google Analytics accounts and properties.
• get_property_details: Returns details about a specific property.
• list_google_ads_links: Returns a list of links to Google Ads accounts for a property.

Core Reports

• run_report: Runs a Google Analytics report using the Data API.
• run_funnel_report: Runs a Google Analytics funnel report using the Data API.
• get_custom_dimensions_and_metrics: Retrieves custom dimensions and metrics for a specific property.

Realtime Reports

• run_realtime_report: Runs a Google Analytics realtime report using the Data API.

To use the server with Gemini CLI or Gemini Code Assist, add the analytics-mcp server to your ~/.gemini/settings.json file under the mcpServers key.

Replace PATH_TO_CREDENTIALS_JSON with the path obtained during the ADC setup and YOUR_PROJECT_ID with your Google Cloud project ID.

{
  "mcpServers": {
    "analytics-mcp": {
      "command": "pipx",
      "args": ["run", "analytics-mcp"],
      "env": {
        "GOOGLE_APPLICATION_CREDENTIALS": "PATH_TO_CREDENTIALS_JSON",
        "GOOGLE_PROJECT_ID": "YOUR_PROJECT_ID"
      }
    }
  }
}

Before using the MCP server, you must enable the following APIs in your Google Cloud project:

1. Google Analytics Admin API
2. Google Analytics Data API

Add the MCP server to Claude Code using the claude mcp add command. Replace PATH_TO_CREDENTIALS_JSON with your credentials path and YOUR_PROJECT_ID with your Google Cloud project ID.

claude mcp add analytics-mcp \
  --scope user \
  -e "GOOGLE_APPLICATION_CREDENTIALS=PATH_TO_CREDENTIALS_JSON" \
  -e "GOOGLE_PROJECT_ID=YOUR_PROJECT_ID" \
  -- pipx run analytics-mcp

The Google Analytics MCP server can be executed as a standalone process using standard I/O (stdio). This is the primary way to interface the server with MCP clients like Claude Code or Gemini.

To run the server, execute the server.py script directly using a Python interpreter. The server communicates via standard input and output streams.

Once configured, you can interact with your Google Analytics data using natural language. Examples include:

• Capability check: what can the analytics-mcp server do?
• Property details: Give me details about my Google Analytics property with 'xyz' in the name
• Data analysis: what are the most popular events in my Google Analytics property in the last 180 days?
• User analysis: were most of my users in the last 6 months logged in?
• Configuration check: what are the custom dimensions and custom metrics in my property?

The call_mcp_tool function is an asynchronous MCP server handler used to execute a specific tool by name with provided arguments. It handles tool execution via the internal tool_map, serializes the response to JSON, and returns it as a list of mcp_types.TextContent. If execution fails, it returns a JSON-formatted error message.

@app.call_tool()
async def call_mcp_tool(name: str, arguments: dict) -> list[mcp_types.Content]:
    # ... implementation details ...

The list_tools function is an asynchronous MCP server handler that returns a list of available mcp_types.Tool objects. These tools include account summaries, property details, and various reporting capabilities (core, realtime, funnel, and conversions).

@app.list_tools()
async def list_tools() -> list[mcp_types.Tool]:
    return mcp_tools

The sanitize_mcp_schema_properties function ensures that the JSON schema for MCP tools is compatible with strict clients (like Claude Desktop). It recursively traverses a schema node and ensures that additionalProperties is set to a boolean value (True) rather than a schema object, which can cause client-side failures.

def sanitize_mcp_schema_properties(node: dict) -> None: