canvas-mcp

canvas-mcp

Connects Claude to Canvas LMS for managing coursework through natural language, including assignments, discussions, submissions, and course browsing.

Category
访问服务器

README

Canvas LMS MCP Server for Claude

An MCP (Model Context Protocol) server that connects Claude to Canvas LMS, letting you manage your coursework through natural conversation.

Ask Claude things like:

  • "What assignments do I have due this week?"
  • "Show me the rubric for my essay assignment"
  • "What did my professor say in the feedback?"
  • "List all my courses"
  • "Are there any overdue assignments?"

Quick Start

1. Get Your Canvas API Token

  1. Log in to Canvas
  2. Click your profile pictureSettings
  3. Scroll to Approved Integrations
  4. Click + New Access Token
  5. Name it (e.g., "Claude") and click Generate Token
  6. Copy the token - you won't see it again!

2. Install the MCP Server

# Clone this repository
git clone https://github.com/lucanardinocchi/canvas-mcp.git
cd canvas-mcp

# Install dependencies
npm install

# Build the project
npm run build

3. Configure Claude Desktop

Open Claude Desktop's config file:

  • macOS: ~/Library/Application Support/Claude/claude_desktop_config.json
  • Windows: %APPDATA%\Claude\claude_desktop_config.json

Add (or merge) this configuration:

{
  "mcpServers": {
    "canvas": {
      "command": "node",
      "args": ["/FULL/PATH/TO/canvas-mcp/dist/index.js"],
      "env": {
        "CANVAS_API_TOKEN": "YOUR_TOKEN_HERE",
        "CANVAS_BASE_URL": "https://your-school.instructure.com"
      }
    }
  }
}

Replace:

  • /FULL/PATH/TO/canvas-mcp with the actual path where you cloned this repo
  • YOUR_TOKEN_HERE with your Canvas API token
  • https://your-school.instructure.com with your Canvas URL

Common Canvas URLs:

University Canvas URL
University of Sydney https://canvas.sydney.edu.au
Generic Instructure https://canvas.instructure.com
Your school Check your browser when logged into Canvas

4. Restart Claude Desktop

Quit Claude Desktop completely (Cmd+Q / Alt+F4) and reopen it.

You should now see Canvas tools available in Claude!


What Can It Do?

📚 Courses

  • List all your enrolled courses
  • Get course details and syllabi

📝 Assignments

  • List assignments (filter by upcoming, overdue, etc.)
  • View full assignment details and instructions
  • See rubrics and grading criteria
  • Check your grades and submission status

💬 Discussions

  • View discussion boards
  • Post new discussion entries
  • Reply to classmates

📤 Submissions

  • Submit text assignments directly
  • Upload files for submission
  • View instructor feedback and comments

🔍 Search

  • Find assignments by due date
  • Search course content
  • Get all upcoming work across all courses

All Available Tools

Tool What it does
list_courses List your enrolled courses
get_course Get details about a specific course
list_assignments List assignments (with filters)
get_assignment Get full assignment details + rubric
get_rubric Get grading rubric for an assignment
get_submission View your submission and feedback
submit_assignment Submit text or URL to an assignment
upload_file Upload a file for submission
list_modules Browse course modules
list_announcements Get course announcements
list_discussions View discussion topics
get_discussion_entries Read discussion posts
post_discussion_entry Post to a discussion
reply_to_discussion Reply to a discussion post
find_assignments_by_due_date Find assignments in a date range
get_upcoming_assignments Get work due in the next N days
get_overdue_assignments Find past-due work
search_course_content Search modules and assignments
get_all_upcoming_work Upcoming work across ALL courses

Troubleshooting

"Canvas tools not showing up"

  1. Make sure you restarted Claude Desktop completely
  2. Check that the path in your config is correct
  3. Verify your claude_desktop_config.json is valid JSON

"401 Unauthorized" errors

Your API token is invalid or expired. Generate a new one in Canvas settings.

"403 Forbidden" errors

You don't have access to that resource. The course may have ended, or you're not enrolled.

"Connection refused" or network errors

Check that your CANVAS_BASE_URL is correct (no trailing slash).


Security Notes

⚠️ Keep your API token secret!

  • Never commit your token to git
  • Don't share your token with others
  • Set an expiration date when creating tokens
  • Revoke tokens you no longer use (Canvas Settings → Approved Integrations)

Your API token has the same access as your Canvas account - anyone with it can view your grades, submit assignments, etc.


Development

# Install dependencies
npm install

# Build once
npm run build

# Watch mode (auto-rebuild on changes)
npm run dev

Project Structure

canvas-mcp/
├── src/
│   ├── index.ts           # MCP server entry point
│   ├── canvas-client.ts   # Canvas API wrapper
│   ├── tools/             # Tool implementations
│   │   ├── courses.ts
│   │   ├── assignments.ts
│   │   ├── submissions.ts
│   │   ├── discussions.ts
│   │   ├── modules.ts
│   │   └── search.ts
│   └── types/
│       └── canvas.ts      # TypeScript types
├── dist/                  # Compiled output
├── package.json
└── tsconfig.json

Contributing

Contributions welcome! Feel free to:

  • Report bugs
  • Suggest new features
  • Submit pull requests

License

MIT - Use it however you want!


Acknowledgments

Built with:

推荐服务器

Baidu Map

Baidu Map

百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。

官方
精选
JavaScript
Playwright MCP Server

Playwright MCP Server

一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。

官方
精选
TypeScript
Magic Component Platform (MCP)

Magic Component Platform (MCP)

一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。

官方
精选
本地
TypeScript
Audiense Insights MCP Server

Audiense Insights MCP Server

通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。

官方
精选
本地
TypeScript
VeyraX

VeyraX

一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。

官方
精选
本地
graphlit-mcp-server

graphlit-mcp-server

模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。

官方
精选
TypeScript
Kagi MCP Server

Kagi MCP Server

一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。

官方
精选
Python
e2b-mcp-server

e2b-mcp-server

使用 MCP 通过 e2b 运行代码。

官方
精选
Neon MCP Server

Neon MCP Server

用于与 Neon 管理 API 和数据库交互的 MCP 服务器

官方
精选
Exa MCP Server

Exa MCP Server

模型上下文协议(MCP)服务器允许像 Claude 这样的 AI 助手使用 Exa AI 搜索 API 进行网络搜索。这种设置允许 AI 模型以安全和受控的方式获取实时的网络信息。

官方
精选