Collaboration

Thanks for your interest in our solution. Having specific examples of replication and usage allows us to continue to grow and scale our work. If you clone or use this repository, kindly shoot us a quick email to let us know you are interested in this work!

wwps-cic@amazon.com

Disclaimers

Customers are responsible for making their own independent assessment of the information in this document.

This document:

Customers are responsible for making their own independent assessment of the information in this document.

This document:

(a) is for informational purposes only,

(b) references AWS product offerings and practices, which are subject to change without notice,

(c) does not create any commitments or assurances from AWS and its affiliates, suppliers or licensors. AWS products or services are provided “as is” without warranties, representations, or conditions of any kind, whether express or implied. The responsibilities and liabilities of AWS to its customers are controlled by AWS agreements, and this document is not part of, nor does it modify, any agreement between AWS and its customers, and

(d) is not to be considered a recommendation or viewpoint of AWS.

Additionally, you are solely responsible for testing, security and optimizing all code and assets on GitHub repo, and all such code and assets should be considered:

(a) as-is and without warranties or representations of any kind,

(b) not suitable for production environments, or on production or other critical data, and

(c) to include shortcuts in order to support rapid prototyping such as, but not limited to, relaxed authentication and authorization and a lack of strict adherence to security best practices.

All work produced is open source. More information can be found in the GitHub repo.

Canvas MCP Server

A local MCP server that lets you manage Canvas LMS — courses, assignments, grades, modules, and more — from any MCP-compatible client like Claude Desktop, Kiro, or Amazon Q Developer.

Quick Start

1. Get a Canvas API Token

1. Log in to your Canvas instance

2. Go to Account → Settings

3. Scroll to Approved Integrations and click + New Access Token

4. Give it a name, click Generate Token, and copy it

2. Install

git clone https://github.com/cal-poly-dxhub/canvas-mcp && cd canvas-mcp
pip install .

3. Run (standalone / local testing only)

If you're using an MCP client like Kiro, Claude Code, or Claude Desktop, skip this step. The client launches the server automatically.

export CANVAS_API_TOKEN="your-token-here"
export CANVAS_BASE_URL="https://myschool.instructure.com"
canvas-mcp

The server is now listening on stdio for MCP messages.

Kiro

Add to your Kiro MCP configuration:

{
  "mcpServers": {
    "canvas": {
      "command": "/path/to/venv/bin/canvas-mcp",
      "env": {
        "CANVAS_API_TOKEN": "your-token-here",
        "CANVAS_BASE_URL": "https://myschool.instructure.com"
      }
    }
  }
}

Run the new agent under kiro-cli as follows

kiro-cli --agent canvas-agent

Note: You do not need to manually export environment variables or start the server. Kiro automatically launches the MCP server and injects the credentials from the env block above.

Claude Code

claude mcp add canvas -- canvas-mcp

Set credentials before launching, since Claude Code inherits environment variables from your shell:

export CANVAS_API_TOKEN="your-token-here"
export CANVAS_BASE_URL="https://myschool.instructure.com"
claude

Related MCP server: Canvas LMS MCP Server

Connect to Your MCP Client

Claude Desktop

Edit ~/Library/Application Support/Claude/claude_desktop_config.json (macOS) or %APPDATA%\Claude\claude_desktop_config.json (Windows):

{
  "mcpServers": {
    "canvas": {
      "command": "canvas-mcp",
      "env": {
        "CANVAS_API_TOKEN": "your-token-here",
        "CANVAS_BASE_URL": "https://myschool.instructure.com"
      }
    }
  }
}

Tip: If you installed into a virtualenv, use the full path to the canvas-mcp binary, e.g. "/path/to/venv/bin/canvas-mcp".

Available Tools

Student

Instructor

Example Workflows

Student: "What's due this week?"

"Show me everything that's due in the next 7 days."

The agent calls: list_upcoming_assignments(days_ahead=7) → returns assignments sorted by due date across all courses

Student: "Did I turn in my homework?"

"Did I submit the Lab 3 assignment for CS 101?"

The agent calls: get_my_courses → list_course_assignments(course_id) → check_submission_status(course_id, assignment_id) → returns submitted_at, late, missing, score

Student: "How am I doing in this class?"

"What are my grades in Math 200?"

The agent calls: get_my_courses → view_my_grades(course_id) → returns all assignment scores, late/missing flags

Student: "What do I need to work on?"

"What's on my Canvas TODO list?"

The agent calls: view_todo_list() → returns unsubmitted assignments and other items needing attention

Student: "Schedule study time for my midterm"

"Block off 2 hours on Saturday afternoon to study for my CS 101 midterm."

The agent calls: schedule_canvas_event(title="CS 101 Midterm Study", start_at="...", end_at="...")

Instructor: Build a course module in one prompt

"Create a 'Week 7 — Neural Networks' module in my CS 301 course with a reading page on backpropagation and a homework assignment worth 50 points due next Friday."

The agent chains: get_my_courses → create_course_module → create_course_page → create_course_assignment → add_item_to_module × 2

Instructor: Check assignment grades

"How did students do on the Week 3 Problem Set in Math 200?"

The agent calls: get_my_courses → list_course_assignments(course_id) → get_assignment_grade_summary → returns avg, high, low, submission count, missing count

Instructor: Find course files

"Find the syllabus in my English 102 course."

The agent calls: get_my_courses → find_course_files(search_term="syllabus")

Instructor: Post an announcement

"Remind my Biology 101 students that the midterm is next Wednesday and to review chapters 5–8."

The agent calls: get_my_courses → post_course_announcement

Requirements

• Python 3.10+

• A Canvas LMS instance with API access

• A valid Canvas API token

License

MIT-0