Jira

A containerized Python MCP server for Cursor to provide access to Jira.

[!IMPORTANT] This project is experimental and was initially created as a learning exercise. Be aware there are more capable and mature Jira MCP solutions available, such assooperset/mcp-atlassian, and Atlassian's ownMCP Server.

See alsoredhat-ai-tools/jira-mcp-snowflakewhich provides another way to access Red Hat Jira data.

- Go toRed Hat Jira Personal Access Tokensand create a token
- Create a .env file similar toexample.env

{ "mcpServers": { "jiraMcp": { "command": "podman", "args": [ "run", "-i", "--rm", "--env-file", "/path/to/your/dotenv/file.env", "quay.io/sbaird/jira-mcp:latest" ], "description": "A containerized MCP server to query Jira issues" } } }

- podman- Install withsudo dnf install podman(Fedora/RHEL) orbrew install podman(macOS)
- yq- Install withsudo dnf install yq(Fedora/RHEL) orbrew install yq(macOS)
- make- Usually pre-installed on most systems

git clone git@github.com:redhat-ai-tools/jira-mcp.git cd jira-mcp

- Build the image & configure Cursor
This also creates a~/.rh-jira-mcp.envfile likethis.

- Go toRed Hat Jira Personal Access Tokensand create a token
- Edit the.rh-jira-mcp.envfile in your home directory and paste in the token

export $(grep -v '^#' ~/.rh-jira-mcp.env | xargs) && python server.py --transport http --port 3075

export $(grep -v '^#' ~/.rh-jira-mcp.env | xargs) && python server.py --transport sse --port 3075

from llama_stack_client import LlamaStackClient client = LlamaStackClient(base_url=LLAMA_STACK_URL) mcp_llama_stack_client_response = client.responses.create( model=LLAMA_STACK_MODEL_ID, input="Tell me about RHAISTRAT-24.", tools=[ { "type": "mcp", "server_url": JIRA_MCP_URL, "server_label": "Jira_tools", "headers": { "Authorization": f"Bearer {JIRA_API_TOKEN}" } } ] )

This MCP server provides the following tools:

- get_jira- Get details for a specific Jira issue by key.
- search_issues- Search issues using JQL

- create_issue- Create a new Jira issue with summary, description, type, priority, and assignee
- update_issue- Update an existing issue's summary, description, priority, or assignee
- delete_issue- Delete a Jira issue (use with caution)

- get_issue_comments- Get all comments for a Jira issue
- add_comment- Add a comment to a Jira issue
- delete_comment- Delete a comment from a Jira issue

- assign_issue- Assign a Jira issue to a user
- unassign_issue- Unassign a Jira issue

- transition_issue- Transition a Jira issue to a new status (e.g., "In Progress", "Done")
- get_issue_transitions- Get available transitions for a Jira issue

- add_issue_labels- Add labels to a Jira issue
- remove_issue_labels- Remove labels from a Jira issue

- list_projects- List all projects
- get_project- Get project details by key
- get_project_components- Get components for a project
- get_project_versions- Get versions for a project
- get_project_roles- Get roles for a project
- get_project_permission_scheme- Get permission scheme for a project
- get_project_issue_types- Get issue types for a project

- list_boards- List all boards
- list_sprints- List sprints for a board
- get_sprint- Get sprint details by ID
- get_sprints_by_name- Get sprints by name for a board, optionally filtered by state

- search_users- Search users by query
- get_user- Get user details by account ID
- get_current_user- Get current user info
- get_assignable_users_for_project- Get assignable users for a project
- get_assignable_users_for_issue- Get assignable users for an issue

- make build- Build the image
- make run- Run the container
- make clean- Clean up the built image
- make cursor-config- Modify~/.cursor/mcp.jsonto install this MCP Server
- make setup- Builds the image, configures Cursor, and creates~/.rh-jira-mcp.envif it doesn't exist

- Confirm thatmake runworks
- Check that the JIRA_API_TOKEN is correct
- Verify the image was built successfully withpodman images jira-mcp
- Go to the "Output" tab in Cursor's bottom pane, choose "MCP Logs" from the drop-down select and examine the logs there
- (MacOS) jiraMcp shows up in Cursor tools section but shows no active tools:

- Edit args section of jiraMcp of yourmcp.jsonfile to include your full path to the.rh-jira-mcp.envfile.
- Example:"~/.rh-jira-mcp.env",to"/Users/your_username/.rh-jira-mcp.env",

- Restart Cursor after configuration changes
- Check Cursor's developer console for error messages
- Verify the Jira URL is accessible from your network

This project is licensed under the MIT License. See theLICENSEfile for details.

MCP server for connecting AI assistants to your own Jira instance