canvas-mcp
Local-first MCP server for Canvas LMS. Stdio transport, no third-party broker, token stays in ~/.canvas.env. Exposes courses, assignments, modules, announcements, grades, planner items, upcoming events, todo list, and bulk file dump as MCP tools.
README
canvas-mcp
<!-- mcp-name: io.github.admin978/canvas-mcp -->
Local-first MCP server for Canvas LMS. Stdio transport, no network round-trips beyond the official Canvas API.
Status: alpha. Single-user, no warranty, API surface may still shift. File issues if it breaks.
Why
Canvas is built for instructors. As a student you get a fragmented UI, no cross-course search, and notifications that arrive late or never. This server exposes the Canvas REST API as MCP tools so you can drive the LMS from any MCP-compatible client (Claude Code, Claude Desktop, etc.).
Architecture
[client] ──stdio──> [server.py] ──https──> [Canvas API]
- Token lives in
~/.canvas.env(chmod 600) - Server runs locally, no third party in the path
- Single file, fully auditable
Tools exposed
list_courses, list_assignments, list_modules, list_announcements, get_page, get_file_info, get_grades, planner_items, upcoming_events, todo.
Setup
1. Mint a Canvas personal access token
In Canvas: Account → Settings → Approved Integrations → + New Access Token. Copy the token shown — it is not retrievable afterwards.
2. Install
From PyPI (recommended):
pip install canvas-local-mcp
Or from source:
git clone https://github.com/admin978/canvas-mcp.git && cd canvas-mcp
python3 -m venv .venv && source .venv/bin/activate
pip install -e .
Then create the env file:
curl -fsSL https://raw.githubusercontent.com/admin978/canvas-mcp/main/.canvas.env.example -o ~/.canvas.env
chmod 600 ~/.canvas.env
# edit ~/.canvas.env: set CANVAS_BASE_URL (institution root, no /api/v1)
# and paste the token into CANVAS_TOKEN
3. Register with your MCP client
Claude Code:
claude mcp add canvas-local -- canvas-local-mcp
Claude Desktop (~/Library/Application Support/Claude/claude_desktop_config.json):
{
"mcpServers": {
"canvas-local": {
"command": "canvas-local-mcp"
}
}
}
Bulk dump
canvas-local-mcp-dump downloads every file the user has access to (course materials, syllabi). Useful for offline indexing.
canvas-local-mcp-dump # all active courses
canvas-local-mcp-dump 12345 67890 # specific course IDs
Output goes to ./canvas-dump/ by default. Override with CANVAS_DUMP_DIR=/path/to/dir.
License
MIT — see LICENSE.
Built by AGENTE 404 S.L. · admin@agente404.com
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.