概要
What is Jira Context MCP?
Jira Context MCP is a Model Context Protocol (MCP) server that integrates Jira’s issue tracking with AI‑assisted IDEs like Cursor. It enables developers to fetch issue details, retrieve assigned tickets, filter by issue type, and automate workflows directly from their editor.
How to use Jira Context MCP?
Install via Smithery (npx -y @smithery/cli install @rahulthedevil/Jira-Context-MCP --client claude) or manually by cloning the repo, running npm install, creating a .env with your Jira credentials, and building (npm run build). Start the HTTP server with npm start or CLI mode with npm run start:cli. Then in Cursor, open Command Palette, choose “Connect to MCP Server”, and enter http://localhost:3000/sse.
Key features of Jira Context MCP
- Fetch detailed information about a specific Jira issue
- Retrieve issues assigned to you in a project
- Filter issues by type (Bug, Story, Epic, etc.)
- List all available projects and issue types
- Retrieve recent ticket changes over a specified period
- Seamless integration with Cursor IDE
Use cases of Jira Context MCP
- Fix a specific bug by fetching issue details and directing Cursor to resolve it
- Work on your assigned issues by retrieving your task list and having Cursor help
- Fix all bugs in a project by filtering by type and resolving them one by one
- Review recent changes in a project to stay updated on modifications
FAQ from Jira Context MCP
What are the prerequisites?
You need Node.js 20.17.0 or higher, a Jira account with API access, and a Jira API token (generated at Atlassian API Tokens).
How do I install the server?
You can install via Smithery automatically for Claude Desktop (npx -y @smithery/cli install @rahulthedevil/Jira-Context-MCP --client claude) or manually by cloning the repository, installing dependencies, creating a .env file, and building the project.
How do I connect it to Cursor?
In Cursor, open the Command Palette (Ctrl+Shift+P or Cmd+Shift+P), type “Connect to MCP Server”, select it, and enter the server URL (default: http://localhost:3000/sse).
What tools are available?
Six tools: /get_issue, /get_assigned_issues, /get_issues_by_type, /get_projects, /get_issue_types, and /get_recent_changes. Each accepts parameters like issueKey, projectKey, or maxResults.
What does the server require at runtime?
It requires Node.js 20.17.0 or higher and a running Jira instance accessible via the configured JIRA_BASE_URL. The server communicates over HTTP (default port 3000) and uses SSE for Cursor integration.
