The YouTrack MCP server acts as a bridge between JetBrains’ project management platform and AI assistants like Claude or VS Code. It allows users to manage tasks, track bugs, and organize team projects using natural language through their favorite AI-powered tools. Instead of manually switching tabs to update a ticket or search for documentation, this server enables an AI to handle those chores directly, making project management feel like a seamless conversation. Moving into its deeper capabilities, this enterprise-grade tool provides comprehensive access to approximately 80% of the YouTrack REST API. It supports complex operations such as managing agile boards and sprints, tracking time logs, and even generating Gantt charts for dependency analysis and critical path identification. For developers and project managers, this means the AI can perform bulk issue updates, search the knowledge base for existing solutions, or navigate complicated project hierarchies without human intervention. What makes this MCP particularly effective for LLM integration is its dynamic configuration system, which automatically syncs with a specific YouTrack instance upon startup. By fetching real-time custom fields like unique "State" values or "Priority" levels, it ensures the AI assistant never hallucinates non-existent statuses or suggests invalid commands. With support for both stdio and SSE protocols, it provides a reliable, strongly-typed environment that allows AI agents to interact with professional workflows safely and accurately.
Category: Notes & Knowledge Management
Tags: agile, issue-tracking, jetbrains, task-management, youtrack
bash # Clone the repository git clone https://github.com/itsalfredakku/youtrack-mcp.git cd youtrack-mcp # Install dependencies npm install # Configure environment variables cp .env.example .env # Edit .env to set YOUTRACK_URL and YOUTRACK_TOKEN # Build the project npm run build # Start the stdio MCP server npm start For hosted usage or ChatGPT custom connectors (SSE): bash npm run start:remote # Runs at http://localhost:3001/mcp/sse ---.env file: - YOUTRACK_URL: Base URL (e.g., https://your-instance.youtrack.cloud) - YOUTRACK_TOKEN: Permanent token (found in Profile → Tokens) - PROJECT_ID: (Optional) Default project shortName - LOG_LEVEL: error/warn/info/debug (Default: info) - CACHE_ENABLED: true/false (Default: true)~/Library/Application Support/Claude/claude_desktop_config.json): json { "mcpServers": { "youtrack": { "command": "node", "args": ["/abs/path/youtrack-mcp/dist/index.js"], "env": { "YOUTRACK_URL": "https://your-instance.youtrack.cloud", "YOUTRACK_TOKEN": "token", "PROJECT_ID": "PRJ" } } } } VSCode (.vscode/settings.json): json { "servers": { "youtrack": { "command": "node", "args": ["./dist/index.js"], "env": { "YOUTRACK_URL": "https://your-instance.youtrack.cloud", "YOUTRACK_TOKEN": "token" } } } } Cline / Generic / Zed: json { "mcpServers": { "youtrack": { "command": "node", "args": ["/abs/youtrack-mcp/dist/index.js"], "env": { "YOUTRACK_URL": "https://your-instance.youtrack.cloud", "YOUTRACK_TOKEN": "token" } } } } ---issues: Create, update, comment, search, query, count, and manage state transitions. - activities: Track global or issue-specific activity and audit trails. - commands: Apply commands to multiple issues with suggestions and silent execution. - search_assist: Context-aware query auto-completion. - saved_queries: Create, list, update, and delete saved searches. - agile_boards: List boards/sprints, assign issues, and track progress. - knowledge_base: Create/update articles and manage hierarchy. - projects: Discovery, metadata, and custom field summaries. - users: Search users, groups, and manage teams. - time_tracking: Log work and generate time reports. - analytics: Generate Gantt charts, identify critical paths, and resource allocation. - custom_fields: Manage fields, bundles, and project-specific fields. - comments: CRUD operations on issue comments. - subscriptions: Manage notification preferences. - auth: Validate token and OAuth2 status. ---analytics tools that can calculate the "Critical Path." An AI can use these tools to identify which sequence of tasks determines the project's end date. It can then proactively warn the team if a task on the critical path is nearing its deadline or is currently blocked. Example: A project lead asks: "Based on current issue dependencies and progress in the 'Infrastructure' project, show me the critical path.…Part of MCP Servers