mcp-pronotes
An MCP server for Pronote that enables users to access French school data like grades, timetables, and homework through AI assistants. It supports both parent and student accounts, including multi-child management and canteen menu retrieval.
README
mcp-pronotes
MCP server for Pronote -- access French school data from any AI assistant.
The first Model Context Protocol server for Pronote (by Index Education), the #1 school management platform in France used by millions of students and parents. Connect any MCP-compatible AI assistant to your child's school data: grades, timetables, homework, absences, averages, and canteen menus.
Features
| Tool | Description |
|---|---|
get_timetable |
Daily or weekly class schedule (lessons, rooms, teachers, cancellations) |
get_grades |
Grades with class average, min, max, and coefficient per subject |
get_homework |
Upcoming homework assignments with due dates and completion status |
get_absences |
Absences, delays, and punishments for a given period |
get_student_info |
Student profile, class, school name, and available periods |
get_averages |
Subject averages with student, class, min, and max values |
get_menus |
School canteen menus for a date range |
All tools support multi-child accounts: pass child_name to select a
specific child, or omit it to default to the first child.
Quick start
1. Install dependencies
pip install pronotepy "mcp>=1.20.0"
Or from the requirements file:
pip install -r requirements.txt
2. Configure credentials
Create a config.json in the project directory:
{
"pronote_url": "https://XXXX.index-education.net/pronote/mobile.parent.html",
"username": "your_username",
"password": "your_password",
"account_type": "parent"
}
Set account_type to "student" if you are logging in with a student account.
Alternatively, use environment variables (they take precedence over
config.json):
export PRONOTE_URL="https://XXXX.index-education.net/pronote/mobile.parent.html"
export PRONOTE_USERNAME="your_username"
export PRONOTE_PASSWORD="your_password"
export PRONOTE_ACCOUNT_TYPE="parent"
3. Register with Claude Code
claude mcp add pronotes -- python /path/to/mcp-pronotes/server.py
Or with environment variables instead of config.json:
claude mcp add pronotes \
-e PRONOTE_URL="https://XXXX.index-education.net/pronote/..." \
-e PRONOTE_USERNAME="your_username" \
-e PRONOTE_PASSWORD="your_password" \
-- python /path/to/mcp-pronotes/server.py
The server communicates over stdio and works with any MCP-compatible client.
Configuration
The server reads configuration from two sources (environment variables override
config.json):
| Source | Key | Description |
|---|---|---|
| config.json | pronote_url |
Full Pronote URL for your school |
| config.json | username |
Your Pronote username |
| config.json | password |
Your Pronote password |
| config.json | account_type |
"parent" or "student" |
| env var | PRONOTE_URL |
Overrides pronote_url |
| env var | PRONOTE_USERNAME |
Overrides username |
| env var | PRONOTE_PASSWORD |
Overrides password |
| env var | PRONOTE_ACCOUNT_TYPE |
Overrides account_type |
Multi-child support
If you are a parent with multiple children on the same Pronote account, every
tool accepts an optional child_name parameter. Pass the child's full name as
it appears in Pronote. When omitted, the server defaults to the first child.
Use the get_student_info tool to list all children and their classes.
Authentication
The server authenticates with a standard Pronote username and password. This is the same login you use on the Pronote website or mobile app.
ENT (Espace Numerique de Travail): If your school uses an ENT portal for authentication, you may need to provide the ENT-specific Pronote URL. Check pronotepy's documentation for ENT configuration details.
QR code login: Not yet supported. QR code / token-based authentication is planned for a future release.
Session management
Pronote sessions expire after approximately 30 minutes. The server handles this automatically: it caches the session for 25 minutes and reconnects transparently when needed. No manual intervention required.
Requirements
Credits
Built with pronotepy by bain3.
License
MIT -- Copyright 2026 HAL-XP
推荐服务器
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。
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。
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。
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。
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。
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。
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。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器
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。