surveymonkey -by-cdata

This MCP server hosts SurveyMonkey data via the CData JDBC Driver and exposes it through a simple MCP interface in read-only mode. The server is intended for use with local clients (e.g., Claude Desktop) that query live data without requiring you to write SQL. After running, the client can discover and call a set of tools built from the MCP server name (the prefix defined in your .prp file). The available tools include: {servername}_get_tables to list tables, {servername}_get_columns to list columns for a table, and {servername}_run_query to execute a SELECT query. In practice, you simply ask the AI to explore the data model or request specific data, and the underlying tools translate those requests into SQL queries against the SurveyMonkey data sources exposed by the JDBC driver. The JSON-RPC examples show how to call these tools via the standard tools/call endpoint. Use the server locally (stdio) and ensure Claude Desktop (or your chosen client) is configured with the appropriate path to the Java runtime and the JAR/PRP files.

Prerequisites:\n- Java (JDK 8+) installed and accessible on the command line.\n- Maven or a Java build tool to compile the MCP server.\n- Access to the CData JDBC Driver for SurveyMonkey (download and install separately per the driver instructions).\n\nInstallation steps:\n1) Clone the repository:\n git clone https://github.com/cdatasoftware/surveymonkey-mcp-server-by-cdata.git\n cd surveymonkey-mcp-server-by-cdata\n\n2) Build the MCP server:\n mvn clean install\n This will produce the JAR file named something like CDataMCP-jar-with-dependencies.jar.\n\n3) Install the CData JDBC Driver for SurveyMonkey following the provider’s instructions on their site (download link in the README).\n\n4) Create the .prp file for your JDBC connection (surveymonkey.prp) with the required properties (Prefix, ServerName, ServerVersion, DriverPath, DriverClass, JdbcUrl, Tables). The README provides a sample configuration.\n\n5) Prepare Claude/Desktop configuration (or your MCP client) by adding an entry in the mcpServers section that points to the built JAR and the .prp file. Example command line would look like:\n java -jar CDataMCP-jar-with-dependencies.jar /path/to/surveymonkey.prp\n\n6) Run the MCP server: you can start it directly with the command above. The server uses stdio, so it must be run on the same machine as clients like Claude Desktop.\n\n7) (Optional) Test the server using Claude Desktop or a JSON-RPC client with calls like source_get_tables, source_get_columns, and source_run_query as shown in the README.

Tips and common considerations:\n- This is a local, read-only MCP server; updates to data require changes in the underlying SurveyMonkey data source or reconfiguring write-capable MCPs.\n- The server runs via stdio; ensure your client app runs on the same machine.\n- The {servername} in tool names is derived from your Prefix in surveymonkey.prp (e.g., surveymonkey_get_tables becomes available).\n- If the client cannot see the MCP server, fully quit and restart the client after starting the Java process.\n- Ensure the JDBC driver is licensed (license steps provided in the README) after installation.\n- When configuring Claude Desktop or similar clients, place the path to the JAR and the .prp in the arguments as demonstrated in the provided configuration examples.\n- If OAuth is used in the data source, you may need to authenticate in the browser during connection setup; follow the driver’s Connection String utility steps.\n- The JSON-RPC examples in the README show the standard payload structure for tool calls like source_get_tables, source_get_columns, and source_run_query.