Sign InGet Started

Getting Started with the MCP Server

Connect your coding agent to Hamster in under a minute.

Overview

The Hamster MCP server uses Streamable HTTP — no API keys required. Add a few lines to your agent's configuration, authenticate once through your browser, and your agent has full access to your project.

How It Works

  1. Add the server — Open your agent's MCP configuration file (usually .mcp.json in your project root, or the tool's global settings) and add:

    {
  "mcpServers": {
    "hamster": {
      "command": "npx",
      "args": ["-y", "mcp-remote", "https://tryhamster.com/mcp"]
    }
  }
}

  2. Authenticate — The first time your agent calls a Hamster tool, it opens your browser and asks you to sign in. After you approve, the agent receives a session token and subsequent requests are automatic.

  3. Start using tools — Your agent can now call any of the 20 Hamster tools. For example, asking "what are my current tasks?" triggers list_tasks and returns your task list as structured data.

Supported Tools

Claude Code, Cursor, Windsurf, and any other MCP-compatible tool can connect. The mcp-remote package bridges Streamable HTTP for tools that don't support it natively — the only requirement is Node.js (for npx).

What Your Agent Can Do

  • Read your project: Browse briefs, check task details, look up plans and documents — the same information you see in the browser.
  • Write back: Create briefs, add tasks, update status, and manage subtasks. Your agent can track its own progress or break work into smaller pieces as it goes.
  • Switch accounts: If you belong to multiple team accounts, the agent can switch between them using switch_account.
  • Permission-scoped: The server uses your personal session. Your agent only sees data you have access to, and team permissions are enforced the same way as in the browser.

Tips

  • Ask your agent to "check my Hamster tasks" or "what's the next task on my brief?" to trigger the relevant tools naturally.
  • If authentication expires, the agent re-opens your browser to re-authenticate. This happens transparently.
  • You can connect the MCP server and the CLI sync at the same time. The MCP server gives your agent live API access for reading and writing; the CLI sync gives it local markdown files for fast, offline-capable reads. They complement each other.

Related