jira-service-management -by-cdata

This MCP server exposes Jira Service Management data through the MCP interface by wrapping the CData JDBC Driver for Jira Service Management. It runs read-only queries against your Jira data and presents tables and columns as relational objects that LLMs can query without requiring SQL. Tools exposed by the server follow the pattern {servername}_get_tables, {servername}_get_columns, and {servername}_run_query, enabling discovery of available data and execution of SELECT statements against Jira tables. For Claude Desktop or other MCP clients, you’ll typically configure a local Claude/CLI config to point to the Java-based MCP jar and the .prp file that defines your JDBC connection, after which you can ask the agent questions like “How many open Jira tickets in project X?” and receive live results.

Usage is designed to be transparent: you don’t need to invoke individual tools yourself for common questions—the client will route natural language questions to the built-in tools as needed. If you do explore programmatic access, you can use the JSON-RPC tool calls shown in the README to call source_get_tables, source_get_columns, or source_run_query, supplying the appropriate table names and SQL queries. This enables you to integrate Jira data into custom workflows or AI prompts while maintaining the read-only guarantees of this MCP server.

Prerequisites:

• Java Runtime (JRE/JDK) installed on the host machine.
• Maven installed (for building the MCP server).
• Access to download the CData JDBC Driver for Jira Service Management and a valid license.
• A Jira Service Management instance you want to query via JDBC.

Setup steps:

1. Clone the repository and build the MCP server:

git clone https://github.com/cdatasoftware/jira-service-management-mcp-server-by-cdata.git
cd jira-service-management-mcp-server-by-cdata
mvn clean install

This will produce CDataMCP-jar-with-dependencies.jar in your target directory.

2. Download and install the CData JDBC Driver for Jira Service Management from: https://www.cdata.com/drivers/jiraservicedesk/download/jdbc

3. License the driver (example steps):

• Locate the driver license utility, typically under the lib folder of the driver installation.
• Run: java -jar cdata.jdbc.jiraservicemanagement.jar --license
• Provide your name, email, and license (or TRIAL).

4. Prepare a .prp file for your JDBC connection (e.g. jira-service-management.prp) using the properties described in the README, including Prefix, ServerName, ServerVersion, DriverPath, DriverClass, and JdbcUrl.

5. Run the MCP server with Claude/CLI client configured:

java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/jira-service-management.prp

6. Point your MCP client (e.g., Claude Desktop) to the running server configuration as described in the README (config file under mcpServers, with the appropriate java path and the JAR + PRP path).

Tips and common considerations:

• The server is read-only at the MCP layer. Write/update/delete operations are not permitted through this MCP server.
• The MCP server communicates via stdio, so ensure the client runs on the same machine or supports stdio-based communication.
• If you encounter issues seeing the MCP server in Claude Desktop, fully quit Claude Desktop and relaunch to refresh the MCP server list.
• Ensure the Jira datasource connection string (JdbcUrl) is tested successfully in the CData driver utility before creating the .prp file.
• If your data source uses OAuth, you may need to authenticate in a browser during the connection setup.
• Keep the CData JDBC Driver license valid; an invalid or expired license will prevent querying.
• The environment can be tailored by editing jira-service-management.prp; for example, adjust DriverPath, JdbcUrl, and the Prefix to match your environment.