What it does
Canvas LMS MCP server for students. Read only Canvas courses, grades, due dates, plus .ics export.
- Connects 在你的电脑上运行
- Account or key Needs CANVAS_API_TOKEN · 1 setting to fill in (CANVAS_BASE_URL)
- Category 游戏、体育、美食与学习
为什么排在这里
列表按使用分(满分 100)排序,由以下部分组成
| 热度 | 0/80 | GitHub 星标或每周下载量:较大的计 60,较小的计 20 |
|---|---|---|
| 新鲜度 | 12/12 | 30 天内有发布或代码更新得满分 |
| 仍在响应 | 0/8 | 上次检查时云端端点有响应 |
| 使用分 | 12/100 |
质量: 基础检查通过 3/3 项(仅展示,不用于排序) · 上升势头: 暂无数据 (只用于“上升中”标签)
安装
选择你使用的 AI 应用,然后复制设置。
用一句话让 AI 帮你安装
请帮我安装 Canvas LMS MCP for students。按照 https://mcp.popnx.com/i/anonymousonyx22-canvas-lms-student-mcp.md 操作,先把命令给我看,等我同意后再运行。粘贴到能在你电脑上工作的 AI(如 Claude Code、Cursor)。它会读取我们的步骤,先给你看命令,等你同意。无需密钥,无需注册。 · 查看 AI 会读取的全文
通过 3/3 项检查 · 在你的电脑上运行 · ✓ 命令已检查 2026年11月14日
第一步:你用哪个 AI?
选择你用的应用,步骤会显示在下方。
可连接 Claude,但需要先准备
需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次) · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
让 AI 帮你安装
把这段话复制到 Claude 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Claude 安装 MCP server Canvas LMS MCP for students,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "canvas-lms-student": { "command": "npx", "args": [ "-y", "canvas-lms-student-mcp" ], "env": { "CANVAS_BASE_URL": "your-value", "CANVAS_API_TOKEN": "your-value" } } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) CANVAS_API_TOKEN, CANVAS_BASE_URL 先保留 "your-value",请我自己把真实值粘贴到文件或应用的输入框里。不要让我在对话里输入密钥。5) 按约定执行,然后确认 Canvas LMS MCP for students 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
安装 Node.js(如果还没有)
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
node -v- 显示版本号(例如 v26.13.1)= 已安装,可跳过这一步
- 显示 "command not found"(Windows 显示 "is not recognized")= 还没安装,或刚装好但需要关掉终端再重新打开
怎么打开终端
Mac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
装好后关掉终端再打开,重新运行同一条命令,应显示版本号:
node -v安装一次,所有需要 Node.js 的工具都能用
注册并创建 Canvas LMS MCP for students 的密钥
打开开发者网站,注册账户并创建 API 密钥,下一步要用。价格和免费方案请看开发者网站。
填写 1 项设置
这些是程序设置,不是密钥,无需注册。按开发者的说明填写。
CANVAS_BASE_URL您学校的 Canvas 地址,例如 https://myschool.instructure.com查看英文原文
Your school's Canvas address, for example https://myschool.instructure.com
说明为自动翻译,变量名保持原样;不确定时请查看原文。
打开电脑版 Claude 应用的设置文件
SettingsDeveloperEdit Config
粘贴这些设置并保存文件
{ "mcpServers": { "canvas-lms-student": { "command": "npx", "args": [ "-y", "canvas-lms-student-mcp" ], "env": { "CANVAS_BASE_URL": "your-value", "CANVAS_API_TOKEN": "your-value" } } } }只把密钥粘贴到应用的输入框里,不要粘贴到聊天中或发给任何人
退出 Claude 再重新打开
试一试
用 Canvas LMS MCP for students 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude 说正在使用 Canvas LMS MCP for students,或回答中出现 Canvas LMS MCP for students 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
ChatGPT 暂时无法连接 Canvas LMS MCP for students
Canvas LMS MCP for students 需要安装在你的电脑上,而 ChatGPT 只能连接在网上运行的服务器
你可以改用这些方式
A · 使用能连接 Canvas LMS MCP for students 的应用
- 需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次) · 仅限电脑上的 Claude 应用(Windows 或 Mac),网页和手机不可用
- 需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
- 需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
- 需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
B · 游戏、体育、美食与学习 类中能连接 ChatGPT 的其他服务器
从同类中使用较多的里挑选,仅限用地址即可连接、无需密钥的
- ShipSwift✓ ChatGPT
- Anime Garden✓ ChatGPT
- Summer Engine✓ ChatGPT
可连接 Claude Code,但需要先准备
需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
让 AI 帮你安装
把这段话复制到 Claude Code 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Claude Code 安装 MCP server Canvas LMS MCP for students,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):claude mcp add --scope user canvas-lms-student --env CANVAS_BASE_URL=your-value --env CANVAS_API_TOKEN=your-value -- npx -y canvas-lms-student-mcp 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) CANVAS_API_TOKEN, CANVAS_BASE_URL 先保留 "your-value",请我自己把真实值粘贴到文件或应用的输入框里。不要让我在对话里输入密钥。5) 按约定执行,然后确认 Canvas LMS MCP for students 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
安装 Node.js(如果还没有)
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
node -v- 显示版本号(例如 v26.13.1)= 已安装,可跳过这一步
- 显示 "command not found"(Windows 显示 "is not recognized")= 还没安装,或刚装好但需要关掉终端再重新打开
怎么打开终端
Mac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
装好后关掉终端再打开,重新运行同一条命令,应显示版本号:
node -v安装一次,所有需要 Node.js 的工具都能用
注册并创建 Canvas LMS MCP for students 的密钥
打开开发者网站,注册账户并创建 API 密钥,下一步要用。价格和免费方案请看开发者网站。
填写 1 项设置
这些是程序设置,不是密钥,无需注册。按开发者的说明填写。
CANVAS_BASE_URL您学校的 Canvas 地址,例如 https://myschool.instructure.com查看英文原文
Your school's Canvas address, for example https://myschool.instructure.com
说明为自动翻译,变量名保持原样;不确定时请查看原文。
打开终端
Mac:按 Command+空格,输入 Terminal · Windows:打开 PowerShell
粘贴这条命令并按 Enter
claude mcp add --scope user canvas-lms-student --env CANVAS_BASE_URL=your-value --env CANVAS_API_TOKEN=your-value -- npx -y canvas-lms-student-mcp按 Enter 前,把“your-value”改成你的密钥
只把密钥粘贴到应用的输入框里,不要粘贴到聊天中或发给任何人
在 Claude Code 中输入 /mcp
应能看到 Canvas LMS MCP for students 在列表中并显示已连接
试一试
用 Canvas LMS MCP for students 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Claude Code 说正在使用 Canvas LMS MCP for students,或回答中出现 Canvas LMS MCP for students 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Anthropic 帮助页面核实 · 你的界面可能因版本略有不同
可连接 Cursor,但需要先准备
需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
让 AI 帮你安装
把这段话复制到 Cursor 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 Cursor 安装 MCP server Canvas LMS MCP for students,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "mcpServers": { "canvas-lms-student": { "command": "npx", "args": [ "-y", "canvas-lms-student-mcp" ], "env": { "CANVAS_BASE_URL": "your-value", "CANVAS_API_TOKEN": "your-value" } } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) CANVAS_API_TOKEN, CANVAS_BASE_URL 先保留 "your-value",请我自己把真实值粘贴到文件或应用的输入框里。不要让我在对话里输入密钥。5) 按约定执行,然后确认 Canvas LMS MCP for students 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
安装 Node.js(如果还没有)
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
node -v- 显示版本号(例如 v26.13.1)= 已安装,可跳过这一步
- 显示 "command not found"(Windows 显示 "is not recognized")= 还没安装,或刚装好但需要关掉终端再重新打开
怎么打开终端
Mac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
装好后关掉终端再打开,重新运行同一条命令,应显示版本号:
node -v安装一次,所有需要 Node.js 的工具都能用
注册并创建 Canvas LMS MCP for students 的密钥
打开开发者网站,注册账户并创建 API 密钥,下一步要用。价格和免费方案请看开发者网站。
填写 1 项设置
这些是程序设置,不是密钥,无需注册。按开发者的说明填写。
CANVAS_BASE_URL您学校的 Canvas 地址,例如 https://myschool.instructure.com查看英文原文
Your school's Canvas address, for example https://myschool.instructure.com
说明为自动翻译,变量名保持原样;不确定时请查看原文。
填写你的密钥
在 Canvas LMS MCP for students 的设置中,CANVAS_API_TOKEN, CANVAS_BASE_URL 还是空的,填入你的密钥并保存
只把密钥粘贴到应用的输入框里,不要粘贴到聊天中或发给任何人
在 Cursor 中以 Agent 模式打开 AI 聊天
试一试
用 Canvas LMS MCP for students 告诉我你能帮我做什么,然后做一件给我看✓ 如果 Cursor 说正在使用 Canvas LMS MCP for students,或回答中出现 Canvas LMS MCP for students 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 Cursor 帮助页面核实 · 你的界面可能因版本略有不同
可连接 VS Code,但需要先准备
需要 Canvas LMS MCP for students 的密钥(在开发者网站创建) · 需要填写 1 项设置(CANVAS_BASE_URL) · 需要先在电脑上安装 Node.js(只需一次)
让 AI 帮你安装
把这段话复制到 VS Code 的对话里。AI 会先检查电脑,告诉你要改哪个文件,并且每次都等你同意。
请在这台电脑上为 VS Code 安装 MCP server Canvas LMS MCP for students,按顺序:1) 用 node -v 检查是否有 Node.js。没有的话,请让我从 https://nodejs.org/en/download 安装,不要自己安装。2) 只使用这份配置(已通过 PopNx MCP 检查):{ "servers": { "canvas-lms-student": { "type": "stdio", "command": "npx", "args": [ "-y", "canvas-lms-student-mcp" ], "env": { "CANVAS_BASE_URL": "your-value", "CANVAS_API_TOKEN": "your-value" } } } } 3) 告诉我要修改的文件完整路径或要运行的命令,等我同意。4) CANVAS_API_TOKEN, CANVAS_BASE_URL 先保留 "your-value",请我自己把真实值粘贴到文件或应用的输入框里。不要让我在对话里输入密钥。5) 按约定执行,然后确认 Canvas LMS MCP for students 的工具真的出现了。这段话不包含你的密钥。如果需要密钥,AI 会请你自己粘贴到文件或应用的输入框里。
安装 Node.js(如果还没有)
已经装好了吗?先检查 · 打开终端,输入这条命令后按 Enter
node -v- 显示版本号(例如 v26.13.1)= 已安装,可跳过这一步
- 显示 "command not found"(Windows 显示 "is not recognized")= 还没安装,或刚装好但需要关掉终端再重新打开
怎么打开终端
Mac:按 ⌘ + Space,输入 Terminal 后按 Enter
Windows:打开“开始”菜单,输入 Windows PowerShell,然后选“打开”
还没安装: 打开官方下载页,选择 LTS 版本和点选式安装程序:Mac 用 .pkg · Windows 用 .msi。双击后一路下一步即可,不需要 curl 命令
装好后关掉终端再打开,重新运行同一条命令,应显示版本号:
node -v安装一次,所有需要 Node.js 的工具都能用
注册并创建 Canvas LMS MCP for students 的密钥
打开开发者网站,注册账户并创建 API 密钥,下一步要用。价格和免费方案请看开发者网站。
填写 1 项设置
这些是程序设置,不是密钥,无需注册。按开发者的说明填写。
CANVAS_BASE_URL您学校的 Canvas 地址,例如 https://myschool.instructure.com查看英文原文
Your school's Canvas address, for example https://myschool.instructure.com
说明为自动翻译,变量名保持原样;不确定时请查看原文。
填写你的密钥
在 Canvas LMS MCP for students 的设置中,CANVAS_API_TOKEN, CANVAS_BASE_URL 还是空的,填入你的密钥并保存
只把密钥粘贴到应用的输入框里,不要粘贴到聊天中或发给任何人
在 VS Code 中以 Agent 模式打开 AI 聊天
试一试
用 Canvas LMS MCP for students 告诉我你能帮我做什么,然后做一件给我看✓ 如果 VS Code 说正在使用 Canvas LMS MCP for students,或回答中出现 Canvas LMS MCP for students 的工具名称,就表示连接成功
菜单名称已于 2026年11月11日 对照 VS Code 帮助页面核实 · 你的界面可能因版本略有不同
开发者专用:终端命令和 JSON 设置
安装在电脑上
Claude Code
claude mcp add --scope user canvas-lms-student --env CANVAS_BASE_URL=your-value --env CANVAS_API_TOKEN=your-value -- npx -y canvas-lms-student-mcpClaude Desktop (claude_desktop_config.json)
{
"mcpServers": {
"canvas-lms-student": {
"command": "npx",
"args": [
"-y",
"canvas-lms-student-mcp"
],
"env": {
"CANVAS_BASE_URL": "your-value",
"CANVAS_API_TOKEN": "your-value"
}
}
}
}Cursor (~/.cursor/mcp.json)
{
"mcpServers": {
"canvas-lms-student": {
"command": "npx",
"args": [
"-y",
"canvas-lms-student-mcp"
],
"env": {
"CANVAS_BASE_URL": "your-value",
"CANVAS_API_TOKEN": "your-value"
}
}
}
}VS Code (mcp.json)
{
"servers": {
"canvas-lms-student": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"canvas-lms-student-mcp"
],
"env": {
"CANVAS_BASE_URL": "your-value",
"CANVAS_API_TOKEN": "your-value"
}
}
}
}需要你填写的值
CANVAS_BASE_URL必填 env您学校的 Canvas 地址,例如 https://myschool.instructure.com查看英文原文
Your school's Canvas address, for example https://myschool.instructure.comCANVAS_API_TOKEN机密 必填 envA Canvas access token, created in Canvas under Account > Settings > New Access Token
说明为自动翻译,变量名保持原样;不确定时请查看原文。
这些设置是根据作者在注册中心的条目生成的。如果 README 的写法不同,请以 README 为准。
一览
哪些应用可以使用它
依据这个服务器的连接方式和各应用自己的文档,检查于 2026年11月10日。
在你的电脑上运行,适用于
ClaudeGooseLM StudioMsty StudioCherry StudioCursorVS Code + GitHub CopilotVisual StudioWindsurf (Devin Desktop)ZedKiroClineContinueJetBrains AI AssistantClaude CodeCodex CLIGemini CLIOpenCodenodecodeWarpLibreChatRaycast
它能做什么
它在你的电脑上运行;安装后,工具列表会显示在你的 AI 应用中。
基础检查i
我们检查的是任何人都能看到的信息;我们没有阅读代码。不适用的检查项不计入总数。 · 检查于 2026年11月19日
- ✓ 通过
作者与仓库一致
注册名称已通过 GitHub 账户 AnonymousOnyx22 验证,该账户拥有此仓库
- ✓ 通过
仍在维护
代码最后更新于18天前
- ✓ 通过
有许可证
使用 MIT
- – 不适用
端点有响应
它在你的电脑上运行,没有互联网端点
- – 不适用
使用 OAuth 登录
在你的电脑上运行,改用本地设置
- … 尚未检查
工具描述中没有隐藏指令
它在你的电脑上运行;我们只能读取互联网端点的工具列表
已公开的漏洞(OSV.dev)
OSV.dev 中没有 canvas-lms-student-mcp@0.1.10 的记录
只检查这个版本的包本身,不包括所有依赖 · 检查于 2026年11月15日
作者:申请重新检查或报告信息错误
给作者:在 README 中加徽章
如果你是作者,可把这段代码贴到 README。徽章会链接回本页。
[](https://mcp.popnx.com/en/servers/anonymousonyx22-canvas-lms-student-mcp)讨论
提问、分享技巧或报告无法使用。按“有帮助”数量排序。
用户评价
还没有评价
与“游戏、体育、美食与学习”中的其他服务器比较i
- 星标数多于此分类中其他 1619 个服务器的 0%
- 发布到注册中心的时间比此分类中其他 1619 个服务器的 58% 更新
- 基础检查结果等于或优于此分类中其他 1619 个服务器的 99%
与注册中心里这个分类中所有活跃的服务器比较。只有当发布者拥有该仓库时,星标数才计入。
常见问题
Canvas LMS MCP for students 安全吗?
它在 3 项适用的基础检查中通过了 3 项。我们没有阅读代码。在添加密钥或令牌之前,请先看看这个服务器要求哪些权限,并使用只具备所需权限的密钥。
它需要哪些密钥或代码?
你需要填写:CANVAS_API_TOKEN
它在我的电脑上运行还是在云端?
它在你的电脑上运行。你的 AI 应用会在需要时启动它,数据不会经过作者的服务器(除非它自己调用在线服务)。
它还在维护吗?
GitHub 仓库最后更新于18天前,最新版本(0.1.10)于18天前进入注册中心。
哪些应用可以使用它?
支持 MCP 的应用,例如 Claude Desktop、Claude Code、Cursor 和 VS Code。你的电脑上还需要有 Node.js。
有什么可以替代它?
同一分类中有:Anime Garden, WilmAI, Lumify Sports Intelligence, Savecraft, GoodWatch
功能相近的其他选择
同一分类、其他作者、命令已通过检查。官方的排在前面,再按星标排序。
如何看卡片:↓ 每周下载量i★ GitHub 星标i三大支柱:评价 · 检查 · 使用情况i更新 = 最新版本发布到注册中心的时间i
WilmAI
Finland's Wilma school system for parents and students: schedule, homework, exams and messages.
Lumify Sports Intelligence
Schedules, scores, odds, splits & explainable AI bet confidence — 8+ sports, free instant key.
Savecraft
Savecraft serves real save game data and expert game knowledge to AI assistants.
GoodWatch
Find movies and TV shows, track watches, and get suggestions based on your taste.
| 本项 Canvas LMS MCP for students | Anime Garden | WilmAI | Lumify Sports Intelligence | Savecraft | GoodWatch | |
|---|---|---|---|---|---|---|
| 运行位置 | 需要安装在电脑上 | 在云端运行,无需安装 | 需要安装在电脑上 | 在云端运行,无需安装 | 在云端运行,无需安装 | 在云端运行,无需安装 |
| 需要密钥 | 是 | 否 | 否 | 是 | 否 | 否 |
| 可在 Claude 中按 + 添加 | 否 | 否 | 否 | 否 | 否 | 否 |
| 基础检查 | 3/3 | 4/4 | 3/3 | 4/4 | 3/4 | 4/4 |
| 许可证 | MIT | AGPL-3.0 | MIT | MIT | Apache-2.0 | MIT |
| 价格 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 | 登记中未列出费用 |
作者的 README
打开 README
按原文从作者的仓库复制。PopNx MCP 没有审查或背书其内容。HTML、图片和脚本已被移除。链接会离开本站。
Canvas LMS MCP for students
Ask your AI what's actually left to do in Canvas.
Canvas LMS MCP is a free, open source MCP server for Claude, Cursor and other AI apps that run a local MCP server. It reads your Canvas courses, quizzes, files, due dates, grades and announcements, and it is read only in Canvas: it never submits, edits or deletes anything there. The one thing it writes is an .ics calendar file on your own computer, and only when you ask it to.
Install and run it with npx, no install step needed:
npx -y canvas-lms-student-mcp
It reads two environment variables: CANVAS_BASE_URL, your school's Canvas address, and CANVAS_API_TOKEN, a token you create in Canvas.
Works with any school that uses Canvas. Not affiliated with Instructure. Free forever, no sign up and no subscription. Your login details only ever go to your own school.
Quick start
- In Canvas, go to Account > Settings, scroll to Approved Integrations and click New Access Token. Copy it.
- Add this to your AI app's MCP config (for Claude Desktop that's
claude_desktop_config.json):
{
"mcpServers": { "canvas": { "command": "npx", "args": ["-y", "canvas-lms-student-mcp"], "env": { "CANVAS_BASE_URL": "https://yourschool.instructure.com", "CANVAS_API_TOKEN": "paste-your-token-here" } } } }
- Restart your AI app and ask something like "what's due this week?"
To check your setup first, run npx canvas-lms-student-mcp doctor with the two environment variables set.
Some schools turn off student access tokens. If you don't see the New Access Token button, your school has disabled it and this server can't connect to your account.
What you can ask
- "What's due this week?"
- "Do I have anything missing or overdue?"
- "What's my grade in each class, and which assignments make it up?"
- "What did my instructors announce lately?"
- "What's new in Canvas since last week?"
- "Show me what's in the Week 3 module for Networking."
- "Which assignments are still due in my Networking 1 course?"
- "Put my Canvas due dates in Google Calendar."
Tools
All 21 tools read from Canvas LMS and none of them can submit, edit or delete anything there. Every tool description includes an example prompt your AI can copy, and the Ask like column repeats those prompts here. The last one, export_due_dates_calendar, writes an .ics file to a folder you name.
Where a tool works on one course, you can name the course instead of looking up its number. Give the Canvas course ID or the course name, for example "Networking 1", and the name is matched against the list get_my_courses shows. A name that fits no course says so and points back at get_my_courses.
| Tool | What it does | Ask like |
|---|---|---|
get_my_courses | Your active courses and their IDs | "What are my Canvas courses this term?" |
get_todo | Your Canvas to-do list | "What does Canvas say I still need to do?" |
get_upcoming_due_dates | Upcoming assignments and events, soonest first, with an optional days_ahead window | "What is due in the next 3 days?" |
get_quizzes | Quizzes and exams with open, due and close times, time limit, attempts used and score | "Which Canvas quizzes and exams are coming up and how long do I get?" |
get_missing_assignments | Past due work you haven't submitted | "What am I missing in Canvas?" |
get_assignments | Assignments for a course by ID or name, with due dates and submission status | "Which Canvas assignments in Networking 1 are overdue?" |
get_my_grades | Current grade for each course, plus the assignment scores and submission status behind it | "What is my grade in each Canvas course and which assignments make it up?" |
get_course_content | Modules and items for a course by ID or name | "Show me the Week 3 module in my Canvas Networking course." |
get_course_modules | Module progress for a course: each module's state and what each item needs from you, done or not | "Which modules do I still have to finish in my Canvas Networking course?" |
get_syllabus | A course's syllabus text, by ID or name | "What does the Canvas syllabus say about grading in Scripting?" |
get_course_people | Teachers and TAs of a course by ID or name, with role and email, and an optional role filter | "Who teaches my Canvas Networking course and how do I email them?" |
get_announcements | Recent announcements, for the courses you name or for all of them | "What did my instructors announce in Canvas lately?" |
get_whats_new | One catch up: recent announcements, newly graded work and unread discussion replies | "What's new in Canvas since last week?" |
get_inbox | Unread Canvas inbox conversations with who they are with, the date and a preview | "Did any instructor message me in Canvas?" |
get_discussions | Discussion topics for a course by ID or name | "What discussions are open in my Canvas Networking course?" |
get_calendar_events | Calendar events between two dates | "What is on my Canvas calendar between 2026-10-01 and 2026-10-31?" |
get_target_grade | What score is needed on remaining work to reach a target percent | "What do I need on the rest of my work in Networking 1 to get an 85 percent?" |
get_assignment_files | The files linked in one assignment's instructions, with name, type, size and a link | "What files come with the Lab 4 assignment in Networking 1?" |
get_submission_feedback | The instructor's comments on your graded submission, with the score and grade | "What did the teacher say about my Lab 4 submission in Networking 1?" |
get_file | One Canvas file by ID. Text, CSV, JSON and HTML files come back as text, other types give the name, type and size | "Open Canvas file 501 and summarize it." |
export_due_dates_calendar | An .ics file with your due dates, ready to import into Google Calendar or Apple Calendar | "Put my Canvas due dates in Google Calendar." |
Calendar export
export_due_dates_calendar turns your Canvas due dates and your Canvas calendar events into an iCalendar file. Name a folder and the file is saved there, for example "Save my Canvas due dates to my Documents folder as an .ics file". Open the file or import it in Google Calendar, Apple Calendar or Outlook.
The file holds the next 30 days and only the work you still owe. Use days_ahead for a longer window, course_id for one course by ID or name, and include_submitted to add the work you already handed in. Each event carries the course name, the points, the submission status, a link back to Canvas and a reminder the day before. Leave the folder out and the calendar text comes back instead, so you can save it under any name you like.
Privacy and security
- Your token stays on your computer. It is only sent to your own school's Canvas address.
- Nothing is sent to any other server by this project.
- Twenty tools are marked read only. The calendar export writes one .ics file in a folder you name, and nothing else.
- Revoke the token any time in Canvas settings.
Troubleshooting
Errors from Canvas and from the connection say what went wrong and what to fix. The most common ones:
| Message | Fix |
|---|---|
| Not signed in | Your token is missing, expired or was revoked. Make a new one in Canvas (Account > Settings > New Access Token), then run npx canvas-lms-student-mcp doctor. |
| Access denied | Your school turned that feature off for students, or the token belongs to a different Canvas account. |
| Not found | Check the course name or ID with get_my_courses. If every request says not found, CANVAS_BASE_URL is wrong. |
| Rate limited | Wait the number of seconds the message says, then try again. |
| Could not reach | The host in CANVAS_BASE_URL does not exist. Check the spelling, for example https://myschool.instructure.com. |
| did not return Canvas API JSON | The address in CANVAS_BASE_URL is not a Canvas site. Use your school's Canvas address with no extra path. |
Development
npm install
npm run build npm test
Tests run against a fake Canvas server, so they need no account or network. The doctor command, the error messages for a bad token, access denied, a rate limit and a reply that is not Canvas, the .ics calendar export and the lookup of a course by name are part of that suite. A metadata test keeps package.json, server.json, glama.json, llms.txt and the server code on one version number, checks that the tool list in the docs still matches the tools the server registers, counts the read only tools from the code so the number in the docs follows the tools, and checks that the registry entry lists the same environment variables the code reads. Pushing to main and opening a pull request run the same build and test, so the badge at the top of this page shows the current result.
The npm package ships two registry files: server.json for the Model Context Protocol registry and glama.json for the Glama directory. Both carry the name, the description, the version, the license, the keywords, the repository and the homepage, so a listing shows what the server does and where the code lives. The server.json entry also names npx as the runtime and lists the two environment variables a listing site should ask for, with CANVAS_API_TOKEN marked as a secret value. It carries the GitHub repository id as well, and that id stays the same when a repository is renamed.
Why Canvas MCP
If you searched for a Canvas MCP server, this is the one for Canvas. It is free, open source and read only: it can look at your courses, due dates, grades and announcements, and it cannot submit, edit or delete anything in Canvas. You sign in with a token you create in your own Canvas account, and you start it with one npx command.
Canvas (Instructure) and Brightspace (D2L) are separate products, so a server for one cannot sign in to the other or read its courses. If your school runs Brightspace instead, use Brightspace MCP (brightspace-d2l-mcp on npm), our Brightspace server, which is also read only. Neither one moves work or grades between the two systems.
Here is what the 21 tools cover on Canvas, so you can check that your question has a tool before you install it.
| Canvas feature | Tool | What you get |
|---|---|---|
| Your courses | get_my_courses | Active courses with their course code and Canvas course ID |
| To do list | get_todo | Assignments and discussions Canvas still lists as to do, with due dates |
| Upcoming work | get_upcoming_due_dates | Assignments and events across all courses, soonest first, with an optional days_ahead window |
| Missing work | get_missing_assignments | Past due assignments you have not submitted, with the course and the day each was due |
| Assignments | get_assignments | Due dates, points and submission status for one course, with an optional bucket filter |
| Grades | get_my_grades | Current percent and letter grade for each course, plus the scored work behind it |
| Modules | get_course_content | Modules and their items for one course |
| Module progress | get_course_modules | Module state, and for each item what Canvas requires and whether you have done it, with only_open |
| Syllabus | get_syllabus | Syllabus text with the HTML removed |
| Instructors and TAs | get_course_people | Teachers and teaching assistants for one course with their email when Canvas shows it, filtered by role |
| Announcements | get_announcements | Recent announcements for all courses or the ones you name |
| What changed | get_whats_new | Announcements, newly graded work and unread discussion replies in on