Canvas MCP is a set of tools that allows your AI agents to interact with Canvas LMS and Gradescope.
- Find relevant resources - Ability to find relevant resources for a given query in natural language!
- Query upcoming assignments - Not only fetch upcoming assignments, but also provide its breakdown for a given course.
- Get courses and assignments from Gradescope - Query your Gradescope courses and assignments with natural language, get submission status, and more!
- Get courses
- Get modules
- Get module items
- Get file url
- Get calendar events
- Get assignments
- and so much more...
Note down the following beforehand:
- Canvas API Key from
Canvas > Account > Settings > Approved Integrations > New Access Token
- Gradescope Email and Password https://www.gradescope.com/
To install Canvas MCP for Claude Desktop automatically via Smithery:
npx -y @smithery/cli install @aryankeluskar/canvas-mcp --client claude
Or, for Cursor IDE to use canvas-mcp with other models:
npx -y @smithery/cli install @aryankeluskar/canvas-mcp --client cursor
Or, for Windsurf:
npx -y @smithery/cli install @aryankeluskar/canvas-mcp --client windsurf
TODO
Create a .env
file in the root directory with the following environment variables:
SNITHERY_API_KEY=your_snithery_api_key
Add the following to your mcp.json
or claude_desktop_config.json
file:
{
"mcpServers": {
"canvas": {
"command": "npx",
"args": [
"-y",
"@smithery/cli",
"run",
"@aryankeluskar/canvas-mcp"
]
}
}
}
Built by Aryan Keluskar :)