brightspace

The Brightspace MCP Server provides a bridge between your AI assistant and D2L Brightspace data. Once configured, your AI client can query grades, upcoming due dates, announcements, and course content, enabling you to ask natural language questions like “What are my assignments due this week?” or “Show me the latest announcements for my CS course.” It supports multiple AI clients such as Claude, ChatGPT, Cursor, and Windsurf by exposing a standard MCP endpoint that your AI can call to fetch Brightspace information. To use it, install the server, authenticate with your Brightspace account, and then point your AI client’s MCP tool to the server’s address. The server handles retrieving data from Brightspace when requested by your AI, so you can focus on asking precise questions rather than navigating the portal manually.

Typical workflows include querying grades across courses, pulling assignment rubrics and due dates, downloading course materials (PDFs, slides, documents), and listing participants or TAs. You can also ask for planning help, such as building study schedules based on due dates. The tool is designed to be used with your AI’s MCP integration, so refer to your specific AI client’s “Add MCP tool” process and point it at the running brightspace-mcp-server instance.

If you’re using Claude, Cursor, Windsurf, or ChatGPT, start by installing and authenticating the server, then use the client configuration steps described in the README to register the Brightspace MCP as an external tool. After setup and authentication, restart your AI client to discover the new Brightspace MCP tool and begin asking questions about your Brightspace data.

Prerequisites:

• Node.js 18+ installed on your machine or server
• Internet access to fetch the MCP server package from npm

Step-by-step installation:

1. Install Node.js 18+ if you haven’t already. Visit https://nodejs.org and download the LTS version for your platform.
2. Install and configure the Brightspace MCP Server using npx. The README shows the standard setup command:
   • Purdue users: npx brightspace-mcp-server setup --purdue
   • Everyone else: npx brightspace-mcp-server setup This wizard will guide you through credentials, MFA, and configuring your AI client’s MCP tool.
3. After the setup completes, run the authentication step if needed to ensure you’re signed in to Brightspace: npx brightspace-mcp-server auth
4. Start using the MCP server with your AI client. Depending on your client, add Brightspace as an MCP tool using the provided commands, e.g. using npx brightspace-mcp-server@latest as the executable in your client’s configuration.
5. If you need to configure manually for Claude Desktop, Cursor, or Windsurf, refer to the examples in the README to paste the appropriate JSON configuration blocks into the client’s settings files or UI.

Notes:

• The server auto-updates to the latest version when your AI client starts a session, as described in the Updates section of the README.
• If you encounter authentication or setup issues, you can rerun: npx brightspace-mcp-server setup
• Config files are typically stored in paths like ~/.brightspace-mcp/config.json; you may edit them directly if needed.

Tips and common issues:

• Credentials stay on your machine; ensure proper file permissions on the config directory (~/.brightspace-mcp/).
• If you see “Not authenticated” errors, re-run npx brightspace-mcp-server auth or npx brightspace-mcp-server setup to refresh tokens.
• If your AI client’s browser-based login times out, you may need to disable headless mode for the login flow (D2L_HEADLESS variable) or run in headed mode as recommended in the Troubleshooting section.
• For WSL or Docker environments, install Chromium dependencies: npx playwright install-deps chromium, since the server may add --no-sandbox in those environments.
• The config location is typically ~/.brightspace-mcp/config.json; you can edit this file directly if you need to adjust settings or credentials manually.