Motion is an MCP server that integrates the Motion task and scheduling platform with AI assistants such as Claude Desktop, Cursor, and VS Code with Continue. It connects directly to Motion v1 APIs, enabling software engineers, project managers, and busy professionals to control their schedules and workloads using conversational prompts. The server provides full coverage across Motion's core features, including workspaces, projects, recurring tasks, individual tasks, schedules, and task comments. To prevent service interruptions, it features an integrated rate-limiting mechanism backed by a local SQLite database that enforces Motion's limit of 12 requests per three-minute window across restarts. Users can query task statuses, modify priorities, assign work, create project structures, and handle calendar schedules directly from their AI chat environment without navigating the Motion web app.
Category: CRM, ERP & E-commerce
Tags: calendar, motion, productivity, scheduling, tasks
claude_desktop_config.json. 3. Add the server entry: json { "mcpServers": { "motion": { "command": "npx", "args": ["-y", "@h3ro-dev/motion-mcp-server"], "env": { "MOTION_API_KEY": "your-api-key" } } } } 4. Restart your AI client to load the tools.Part of MCP Servers
You can install it globally using npm install -g @h3ro-dev/motion-mcp-server or run it directly using npx inside your client configuration file. You will need a Motion API key from your Motion account settings added to the environment variables under MOTION_API_KEY.
It provides tools to manage tasks, recurring tasks, projects, schedules, comments, users, and workspaces. You can create, list, update, move, or delete tasks, archive projects, and track workspace resources directly through conversational prompts in an MCP-compatible client.
The server works with any Model Context Protocol compatible client, including Claude Desktop on macOS, Windows, and Linux, Cursor, and VS Code using the Continue extension.
Motion enforces a limit of 12 requests per 3 minutes. The server automatically manages this using a local SQLite database to persist request tracking across server restarts and provides informative wait times if the limit is reached.
Yes, it is open-source software released under the MIT license and maintained on GitHub by h3ro-dev.