jira

This Jira MCP server provides a natural language interface to manage Jira projects, issues, and workflows. It exposes a set of 11 tools that allow you to fetch issue details, search with JQL, create and update issues, add comments, retrieve project information, and manage issue transitions. With the built-in natural language capabilities, you can ask questions or give commands like "Get details for PROJ-12345", "Create a Bug in PROJ with summary 'Cannot log in'", or "Show me all open issues in PROJ". The server authenticates requests via API tokens, ensuring secure access to your Jira data. To use it with an MCP client, configure the client to launch the server via Python and point it to server.py, supplying your Jira credentials via environment variables.

Prerequisites:

• Python 3.8 or higher installed on the host
• Jira account with API access and an API token
• MCP-compatible client (Cursor, VS Code, etc.)

1. Clone the repository or download the Jira MCP server:

• git clone <repository-url>/jira-mcp-server
• cd jira-mcp-server

2. Install Python dependencies:

• python3 -m pip install -r requirements.txt

3. Configure credentials:

• Create a .env file in the project root with: JIRA_SERVER=https://your-company.atlassian.net JIRA_EMAIL=your-email@company.com JIRA_API_TOKEN=your-api-token

4. Test the connection (optional but recommended):

• python3 test_connection.py

5. Run the MCP server locally:

• python3 server.py

6. If integrating with an MCP client, add a server configuration similar to:

• For Cursor or VS Code MCP integration, specify the full path to server.py and pass the env vars mentioned above. Ensure the type is stdio and use the absolute path to server.py.

Tips and notes:

• Never commit or share your .env file containing Jira credentials; add it to .gitignore.
• Use Jira API tokens rather than passwords for authentication.
• Ensure the MCP client configuration uses the absolute path to server.py and sets the required environment variables.
• If you run inside a virtual environment, point Python to the interpreter within the venv (e.g., "/path/to/venv/bin/python3").
• Common issues include authentication errors due to incorrect API tokens or Jira URL, and misconfigured MCP client paths. The MCP config must include type: stdio in the client settings.
• Restart the MCP client after changing environment variables or server config.