Example prompts
Show tickets assigned to me with links and a summary of progress.
What's the link and status of that ticket with the new button feature?
Add a comment to PROJ-456 explaining the fix in my current branch.
Add this spreadsheet of user stories as story issues to PROJ. Make the stories medium priority with a due date of 2 weeks from now.
Get started
Quickstart
Install and configure Jira MCP in minutes.
Configure your AI tool
Set up Jira MCP with your preferred AI assistant.Cursor
Configure Cursor IDE with Jira MCP.
Claude Code
Set up Claude Code CLI with Jira MCP.
Windsurf
Configure Windsurf with Jira MCP.
How it works
1
Prompt your AI assistant
You enter questions or commands to an LLM client such as Claude Desktop, Cursor, Windsurf, or ChatGPT.
2
LLM analyzes available tools
The LLM analyzes the available MCP tools and decides which one(s) to use. The
LLM has context of each tool and what it is meant for in human language.
3
MCP server executes commands
The client executes the chosen tool(s) through the MCP server. The MCP server
runs locally on your machine or remotely via an endpoint.
4
Results returned to LLM
The results are sent back to the LLM, which formulates a natural language response and displays data or performs actions using the MCP server.
Architecture
MCP follows a client-server architecture where an MCP host (an AI application like Cursor or Claude Desktop) establishes connections to one or more MCP servers. The MCP host accomplishes this by creating one MCP client for each MCP server. Each MCP client maintains a dedicated connection with its corresponding MCP server.Learn more about MCP architecture
Read the official MCP documentation for details on architecture.