Canvas LMS (Local-First)
@admin978
About Canvas LMS (Local-First)
Local-first MCP server for Canvas LMS. Stdio transport, token stays on your machine, no third-party broker.
Config
Add this server to your MCP-compatible client using the configuration below.
{
"mcpServers": {
"canvas-local": {
"command": "canvas-local-mcp"
}
}
}Tools
No tools detected
We auto-extract tools from the README. The maintainer can list them under a ## Tools heading to populate this section.
Overview
What is Canvas LMS (Local-First)?
Canvas LMS (Local-First) is a local-first MCP server that exposes the Canvas REST API as MCP tools. It is built for instructors who want to drive the LMS from any MCP-compatible client, such as Claude Code or Claude Desktop. The API token is stored in a local ~/.canvas.env file, and the server runs as a single auditable file.
How to use Canvas LMS (Local-First)?
Install the package with pip install canvas-local-mcp. Set your Canvas API token in ~/.canvas.env with permissions chmod 600. Then configure your MCP client (e.g., Claude Desktop) using the command canvas-local-mcp as the server command.
Key features of Canvas LMS (Local-First)?
- Exposes 10 Canvas REST API endpoints as MCP tools.
- Includes a bulk dump script for offline course indexing.
- Runs locally as a single auditable file.
- Uses token-based authentication via a local env file.
- Compatible with any MCP client (Claude Code, Claude Desktop).
Use cases of Canvas LMS (Local-First)?
- Instructors can cross-search course materials without using the Canvas UI.
- Bulk download course content for offline indexing and querying.
- Retrieve upcoming assignments,
Basic information
More Other MCP servers
Awesome Mlops
visengerA curated list of references for MLOps
Production-ready MCP integrations for AI applications
Klavis-AIKlavis AI: MCP integration platforms that let AI agents use tools reliably at any scale
ICSS
chokcoco不止于 CSS
Website
FunnyWolfAdversary simulation and Red teaming platform with AI
FastMCP v2 🚀
jlowin🚀 The fast, Pythonic way to build MCP servers and clients.
Comments