CData MCP Gateway for Jira (Read-Only)

This repository furnishes a dedicated Model Context Protocol (MCP) gateway that facilitates read-only access to your active Jira deployment.

:warning: This release is strictly for querying existing data. For comprehensive data manipulation capabilities (CRUD operations, action execution) and a more streamlined integration process, please explore the complimentary CData MCP Server for Jira (Beta).

Core Objective

The primary function of this read-only MCP server is to enable advanced AI agents, such as Claude Desktop, to interact with current data residing in Jira. This connectivity is achieved by leveraging the robust CData JDBC Driver for Jira.

The underlying CData JDBC Driver abstracts Jira's native structure, presenting it as a conventional relational SQL model. Our server component then wraps this driver, exposing the Jira information via the MCP interface. This permits AI systems to retrieve up-to-date intelligence simply by formulating conversational questions, completely bypassing the need for explicit SQL scripting.

Implementation Roadmap

Follow these steps to deploy and configure the connection:

1. Acquire Source Code: bash git clone https://github.com/cdatasoftware/jira-mcp-server-by-cdata.git cd jira-mcp-server-by-cdata
2. Compile Artifacts: bash mvn clean install This action generates the necessary executable JAR: CDataMCP-jar-with-dependencies.jar
3. Driver Acquisition: Secure the CData JDBC Driver for Jira from the official distribution point: https://www.cdata.com/drivers/jira/download/jdbc
4. License Validation: Activate the JDBC Driver:
   • Locate the driver installation directory (e.g., C:\Program Files\CData\CData JDBC Driver for Jira\ on Windows, or /Applications/CData JDBC Driver for Jira/ on Unix-like systems).
   • Execute the licensing utility: bash java -jar cdata.jdbc.jira.jar --license
   • Input your credentials and use "TRIAL" or your actual license key.
5. Connection String Definition:
   • Launch the Connection String configuration tool: bash java -jar cdata.jdbc.jira.jar
   • Configure your Jira connection parameters (e.g., OAuth flow if applicable) and verify connectivity.
   • Capture the resulting JDBC connection string upon successful validation.
6. Configuration File (.prp) Generation: Create a property file (e.g., jira.prp) detailing the server parameters:
   • Prefix: Identifier prefix for exposed tools.
   • ServerName: Arbitrary name for the service instance.
   • ServerVersion: Version identifier.
   • DriverPath: Absolute path to the downloaded JDBC driver JAR.
   • DriverClass: Fully qualified name of the driver class (cdata.jdbc.jira.JIRADriver).
   • JdbcUrl: The connection string copied in the previous step.
   • Tables: (Optional) Specify particular tables; leave empty to enable access to all accessible entities. env Prefix=jira ServerName=CDataJIRA ServerVersion=1.0 DriverPath=PATH\TO\cdata.jdbc.jira.jar DriverClass=cdata.jdbc.jira.JIRADriver JdbcUrl=jdbc:jira:InitiateOAuth=GETANDREFRESH; Tables=

Integrating with LLM Clients (e.g., Claude Desktop)

1. Client Configuration Update: Modify or create the client's configuration file (e.g., claude_desktop_config.json) to register this new MCP endpoint within the mcpServers block.

   Windows Host: json { "mcpServers": { "{classname_dash}": { "command": "PATH\TO\java.exe", "args": [ "-jar", "PATH\TO\CDataMCP-jar-with-dependencies.jar", "PATH\TO\jira.prp" ] }, ... } }

   Linux/Mac Host: json { "mcpServers": { "{classname_dash}": { "command": "/PATH/TO/java", "args": [ "-jar", "/PATH/TO/CDataMCP-jar-with-dependencies.jar", "/PATH/TO/jira.prp" ] }, ... } } Ensure this configuration file is placed in the client's expected location (e.g., %APPDATA%\Claude\claude_desktop_config.json on Windows). 2. Client Restart: Completely terminate and relaunch the host application (Claude Desktop) for the newly defined MCP server to initialize.

Standalone Server Execution

To operate the gateway independently (useful for testing):

```bash java -jar /PATH/TO/CDataMCP-jar-with-dependencies.jar /PATH/TO/jira.prp

Note: This server operates over standard input/output (stdio), restricting its use to client applications running on the same host machine.

Operational Guidance

Once the gateway is established and recognized by the AI client, users can query the live Jira system conversationally. The client automatically invokes the necessary underlying tools based on the natural language request. For instance:

• "Determine the volume of unresolved defects logged within the 'System Migration' project."
• "List all tasks assigned to me that are scheduled for completion this week."

Available Functions

Tool names utilize the server prefix defined in the .prp file (e.g., jira_):

• {servername}_get_tables: Fetches a delimited list of accessible data entities. Follow this with {servername}_get_columns to inspect schema details. Output is CSV format.
• {servername}_get_columns: Retrieves the field definitions for a specified entity. Output is CSV format.
• {servername}_run_query: Executes custom SQL SELECT statements against the Jira data model.

JSON-RPC Interaction Samples (For Direct Scripting)

These examples adhere to the JSON-RPC 2.0 standard, applicable when bypassing the AI interface for programmatic calls:

Querying Available Entities (jira_get_tables)

{
    "jsonrpc": "2.0",
    "id": 1,
    "method": "tools/call",
    "params": {
        "name": "jira_get_tables",
        "arguments": {}
    }
}

Inspecting Entity Schema (jira_get_columns)

{
    "jsonrpc": "2.0",
    "id": 2,
    "method": "tools/call",
    "params": {
        "name": "jira_get_columns",
        "arguments": {
            "table":  "Issue" 
        }
    }
}

Executing a Read Query (jira_run_query)

{
    "jsonrpc": "2.0",
    "id": 3,
    "method": "tools/call",
    "params": {
        "name": "jira_run_query",
        "arguments": {
            "sql":  "SELECT summary, statusName FROM Issue WHERE projectKey = 'DEV' AND priority = 'Highest'"
        }
    }
}

Debugging Common Issues

1. Server Not Appearing: Verify that the host client application (e.g., Claude Desktop) was completely shut down and restarted after modifying its configuration file.
2. Data Retrieval Failures: Double-check the connection string embedded in the .prp file using the Connection String Utility from Step 5.
3. Connection Errors: If the underlying Jira connection fails, seek assistance from the CData Support Team.
4. General MCP Feedback: For issues related to the server operation itself, engage with the CData Community.

Licensing Details

This server implementation is distributed under the permissive MIT License. Review the accompanying LICENSE file for complete terms regarding use, modification, and distribution.

Supported Data Source Ecosystem

This driver infrastructure supports a vast catalog of endpoints, including (but not limited to): Access, Act CRM, Active Directory, Acumatica, Amazon S3, Asana, Azure DevOps, Basecamp, BigQuery, Box, Confluence, Couchbase, Dropbox, Dynamics 365, Elasticsearch, GitHub, Gmail, Google Analytics, Google Sheets, HubSpot, HDFS, Hive, IBM Informix, JDBC-ODBC Bridge, Jira, Jira Assets, Jira Service Management, Kafka, LDAP, MongoDB, MySQL, NetSuite, OData, Oracle, PostgreSQL, QuickBooks Online, Salesforce, SAP, ServiceNow, SharePoint, Shopify, Slack, Snowflake, SQL Server, Stripe, Teradata, Trello, Workday, Zendesk, and many others. Refer to the complete list in the repository for comprehensive coverage.