Skip to main content
Glama

Canvas MCP Server

by plyght

Canvas MCP Server

A Model Context Protocol server providing comprehensive access to Canvas LMS for AI assistants.

Install MCP Server

Features

Access Canvas LMS functionality through 25 MCP tools:

  • Courses: List and view course information

  • Assignments: Browse, view details, check submissions, and submit work

  • Grades: View grades for individual courses or across all courses

  • Messaging: Send and receive conversations with instructors and peers

  • Calendar: View events and upcoming deadlines

  • To-Do Lists: Track pending tasks and assignments

  • Modules: Navigate course content structure

  • Announcements: Read course and institutional announcements

  • Files: Access and download course materials

  • Quizzes: View quizzes and submissions

  • Users: Search for classmates and instructors

Installation

Prerequisites

  • Bun runtime

  • Canvas LMS account with API access token

Setup

  1. Install dependencies:

bun install
  1. Get your Canvas API token:

    • Log into Canvas

    • Navigate to Account → Settings

    • Scroll to "Approved Integrations"

    • Click "+ New Access Token"

    • Generate and copy the token

  2. Configure environment variables:

export CANVAS_BASE_URL="https://your-institution.instructure.com" export CANVAS_ACCESS_TOKEN="your_access_token_here"
  1. Configure your MCP client with the following settings:

{ "mcpServers": { "canvas": { "command": "bun", "args": ["run", "/absolute/path/to/canvas/src/index.ts"], "env": { "CANVAS_BASE_URL": "https://your-institution.instructure.com", "CANVAS_ACCESS_TOKEN": "your_access_token_here" } } } }

Usage

After setup, interact with Canvas through your MCP client:

Show me all my current courses What assignments do I have due soon? What are my grades in all my classes? Show me my recent Canvas messages

Development

Quality Gates

bun run format # Format code with Prettier bun run lint # Lint code with ESLint bun run typecheck # Type check with TypeScript bun test # Run tests

Build

bun run build # Build for production bun run dev # Run in development mode

Documentation

Architecture

Built with TypeScript and the MCP SDK, this server provides a type-safe interface to the Canvas LMS REST API. All operations use Bearer token authentication and respect Canvas rate limits.

Project Structure

canvas/ ├── src/ │ ├── canvas/ │ │ ├── client.ts # Canvas API client │ │ └── client.test.ts # Client tests │ └── index.ts # MCP server implementation ├── docs/ # Documentation ├── package.json # Dependencies and scripts └── tsconfig.json # TypeScript configuration

Security

  • Store credentials in environment variables

  • Never commit access tokens to version control

  • Use token expiration dates

  • Rotate tokens regularly

  • Follow your institution's API usage policies

API Reference

This server implements the Canvas LMS REST API v1. For detailed API documentation, visit the Instructure Developer Portal.

License

This project uses permissive open-source dependencies with no telemetry or tracking.

-
security - not tested
F
license - not found
-
quality - not tested

remote-capable server

The server can be hosted and run remotely because it primarily relies on remote services or has no dependency on the local environment.

Enables AI assistants to interact with Canvas LMS through 25 comprehensive tools. Supports course management, assignments, grades, messaging, calendar events, and file access through natural language.

  1. Features
    1. Installation
      1. Prerequisites
      2. Setup
    2. Usage
      1. Development
        1. Quality Gates
        2. Build
      2. Documentation
        1. Architecture
          1. Project Structure
        2. Security
          1. API Reference
            1. License

              MCP directory API

              We provide all the information about MCP servers via our MCP API.

              curl -X GET 'https://glama.ai/api/mcp/v1/servers/plyght/canvas-mcp'

              If you have feedback or need assistance with the MCP directory API, please join our Discord server