French Tax MCP Server
Enables AI assistants to calculate French individual income tax and retrieve current tax brackets using official government data. Supports household composition calculations and provides up-to-date tax information for French residents.
README
French Tax MCP Server
An MCP (Model Context Protocol) server that provides French individual income tax calculations to AI assistants.
Current Functionality
This MCP server currently supports:
- Individual Income Tax Calculations: Calculate French income tax (impôt sur le revenu) based on net taxable income and household composition (quotient familial)
- Tax Brackets: Retrieve current French income tax brackets from official government sources
- Dynamic Data: Uses web scraping from service-public.fr to get up-to-date tax information with fallback to hardcoded data
Data Sources
The current version uses web scraping with MarkItDown to fetch tax information from official French government websites (primarily service-public.fr). In future versions, this may be replaced with official APIs or other more reliable data sources when available.
Installation
# Install via pip
pip install french-tax-mcp
# Or install via uv (recommended)
uv pip install french-tax-mcp
MCP Configuration
Add to your MCP configuration file (~/.config/mcp/mcp.json or workspace .kiro/settings/mcp.json):
{
"mcpServers": {
"french-tax-mcp": {
"command": "uvx",
"args": ["french-tax-mcp@latest"],
"env": {
"FASTMCP_LOG_LEVEL": "ERROR"
},
"disabled": false,
"autoApprove": []
}
}
}
Available Tools
calculate_income_tax
Calculate French individual income tax.
Parameters:
net_taxable_income: Net taxable income in euroshousehold_parts: Number of household parts (quotient familial) - defaults to 1.0year: Tax year (optional, defaults to current year)
Example:
Calculate income tax for 50,000€ salary with 2 children
get_tax_brackets
Retrieve current French income tax brackets.
Parameters:
year: Tax year (optional, defaults to current year)
Example:
What are the current French tax brackets?
Usage Examples
Basic calculation:
How much income tax will I pay on 45,000€ salary?
Family situation:
Calculate tax for married couple earning 60,000€ with one child
Tax brackets:
Show me the 2024 French tax brackets
Limitations
- Currently supports only individual income tax calculations for French residents
- Web scraping may occasionally fail (fallback data is used in such cases)
- Tax calculations are for informational purposes only
TODO - Future Features
The following features may be added in future versions:
- [ ] LMNP (Location Meublée Non Professionnelle) - Furnished rental tax calculations
- [ ] Pinel Investment - Real estate investment tax benefits
- [ ] Micro-Enterprise/Auto-Entrepreneur - Business tax calculations
- [ ] Corporate Tax - Company tax calculations
- [ ] Property Tax - Real estate tax information
- [ ] Social Charges - Social security contributions
- [ ] Tax Forms - Detailed form guidance and filling assistance
- [ ] Official API Integration - Replace web scraping with official government APIs
Development
Local Development
# Clone the repository
git clone https://github.com/your-username/french-tax-mcp.git
cd french-tax-mcp
# Install in development mode
pip install -e ".[dev]"
# Run the server locally
python -m french_tax_mcp.server --port 8888
Running Tests
# Run tests
python -m pytest tests/
Legal Notice
This tool provides information for informational purposes only and does not constitute professional tax advice. For advice tailored to your personal situation, please consult a certified public accountant or tax advisor.
Tax information is sourced from official French government websites but may not reflect the most recent changes in tax laws. Always verify calculations with official sources.
License
This project is licensed under the Apache License 2.0 - see the LICENSE file for details.
Links
- Official French Tax Website: https://www.impots.gouv.fr
- MCP Protocol: https://modelcontextprotocol.io/
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。