adobe-analytics -by-cdata

This MCP server exposes Adobe Analytics data via a read-only MCP interface using the CData MCP framework. It wraps the CData JDBC Driver for Adobe Analytics and presents data as relational models that can be queried through the MCP tools exposed by the server. Clients (including LLMs like Claude Desktop) can discover available tables, inspect columns, and run read-only SQL queries against Adobe Analytics without writing SQL directly in many cases. The server runs locally on your machine and uses standard JSON-RPC tool commands to list tables, fetch columns, and execute queries through the built-in tools named after the server (e.g., adobe_analytics_get_tables, adobe_analytics_get_columns, adobe_analytics_run_query).

To use it, configure Claude Desktop (or a similar MCP client) with a config entry that points to the Java process running the CDataMCP JAR and the Adobe Analytics .prp configuration file you created. Once started, you can ask the AI client to navigate the available tables and fields, then issue natural language questions such as "What is the monthly revenue by channel?" or let the client transform questions into appropriate tool calls. Note that this server is read-only in this package, so write/update/delete actions are not available.

Prerequisites:

• Java Runtime Environment (JRE) or JDK installed and available on your PATH
• Maven (for building the MCP server from source)

Installation steps:

1. Clone the repository and navigate into it: git clone https://github.com/cdatasoftware/adobe-analytics-mcp-server-by-cdata.git cd adobe-analytics-mcp-server-by-cdata

2. Build the server jar: mvn clean install This creates the JAR file named CDataMCP-jar-with-dependencies.jar (path may vary depending on the project structure).

3. Install the CData JDBC Driver for Adobe Analytics and license it following the provided instructions from CData:

   • Download the driver from: https://www.cdata.com/drivers/adobeanalytics/download/jdbc
   • License the driver as described in the README (license command and prompt).

4. Create a .prp file for your JDBC connection (e.g., adobe-analytics.prp) with the required properties such as Prefix, ServerName, ServerVersion, DriverPath, DriverClass, and JdbcUrl as shown in the README example.

5. Run the MCP server using the generated jar and the .prp file: java -jar CDataMCP-jar-with-dependencies.jar /PATH/TO/adobe-analytics.prp

6. Configure Claude Desktop (or your MCP client) to point to the running server using the provided command and PRP path in the mcpServers config.

Tips and common aspects:

• The server operates via stdio, so it is intended for local clients on the same machine.
• If Claude Desktop cannot see the MCP server, fully quit and restart Claude Desktop after adding or editing the config.
• Ensure the path to CDataMCP-jar-with-dependencies.jar and adobe-analytics.prp is correct in your startup command.
• The provided tool names follow the pattern {servername}_get_tables, {servername}_get_columns, and {servername}_run_query. Replace {servername} with the actual server key used in your config (e.g., adobe_analytics).
• Since this is a read-only setup, any attempt to perform write/update/delete operations will be blocked by the driver/server configuration.
• If you switch data sources or drivers, update the JdbcUrl and DriverPath in the .prp file accordingly and restart the MCP server.