clarity-mcp-server

# Microsoft Clarity Data Export MCP Server This is a Model Context Protocol (MCP) server for the Microsoft Clarity data export API. It allows you to fetch analytics data from Clarity using Claude for Desktop or other MCP-compatible clients. ## Features - Query Microsoft Clarity analytics data through a simple interface - Filter by up to 3 dimensions (Browser, Device, Country/Region, OS, etc.) - Retrieve various metrics (Scroll Depth, Engagement Time, Traffic, etc.) - Seamlessly integrates with Claude for Desktop and other MCP clients ## Setup and Installation ### Prerequisites - Node.js v16 or higher - A Microsoft Clarity account and API token - Any MCP-compatible client (Claude for Desktop, etc.) ### Installation #### Option 1: Install via npm (recommended) You can install and run this package directly using npm: ```bash # Install globally npm install -g @microsoft/clarity-mcp-server # Run the server clarity-mcp-server ``` #### Option 2: Run with npx without installing You can run the server directly using npx without installing: ```bash npx @microsoft/clarity-mcp-server ``` With either option, you can provide your Clarity API token using the `--clarity_api_token` parameter: ```bash npx @microsoft/clarity-mcp-server --clarity_api_token=your-token-here ``` #### Option 3: Manual Installation 1. Clone or download this repository 2. Install dependencies: ``` npm install ``` 3. Build the TypeScript code: ``` npm run build ``` 4. Run the server: ``` npm start ``` #### Option 4: Install from Claude's gallery To install the Microsoft Clarity MCP Server extension in Claude’s DXT environment: 1. Open **Claude Desktop**. 2. Navigate to **File → Settings → Extensions**. 3. Search for **Microsoft Clarity MCP Server**. 4. Click **Install** to add the extension. 5. Configure your API Token. ## Configuration You can provide the [Clarity data export API](https://learn.microsoft.com/en-us/clarity/setup-and-installation/clarity-data-export-api) token in two ways: 1. **Command Line Arguments**: ```bash npx @microsoft/clarity-mcp-server --clarity_api_token=your-token ``` 2. **Tool Parameters**: - Provide `token` as a parameter when calling the `get-clarity-data` tool ## Configuring MCP Clients ### Generic MCP Client Configuration MCP clients typically require configuration to connect to the server. Here's a general example of how to configure an MCP client: ```json { "mcpServers": { "@microsoft/clarity-mcp-server": { "command": "npx", "args": [ "@microsoft/clarity-mcp-server", "--clarity_api_token=your-api-token-here" ] } } } ``` The specifics of where and how to add this configuration will depend on your specific MCP client. ### Claude for Desktop Configuration To configure Claude for Desktop to use this server: 1. Open your Claude for Desktop configuration file: - **Windows**: `%AppData%\Claude\claude_desktop_config.json` - **macOS**: `~/Library/Application Support/Claude/claude_desktop_config.json` 2. Add the configuration shown in the generic example above 3. Save the configuration file and restart Claude for Desktop ## Using the Server When using an MCP client with this server configured, you can ask it to fetch Clarity data. For example: "Can you fetch my Clarity data for the last day, filtered by Browser and showing Traffic metrics?" The MCP client will then prompt you to run the `get-clarity-data` tool, which requires: - `numOfDays`: Number of days to retrieve (1-3) - `dimensions`: Array of dimensions to filter by (optional) - `metrics`: Array of metrics to retrieve (optional) If you haven't configured your credentials via command-line arguments, you'll also need to provide: - `token`: Your Clarity API token ## API Token ### Getting Your API Token To generate an API token: 1. Go to your Clarity project 2. Select Settings → Data Export → Generate new API token 3. Provide a descriptive name for the token 4. Save the generated token securely ## Limitations - Maximum of 10 API requests are allowed per project per day - Data retrieval is confined to the previous 1 to 3 days - Maximum of three dimensions can be passed in a single request - The response is limited to 1,000 rows and can't be paginated ## License MIT