estv-tax
Enables AI assistants to calculate Swiss income, wealth, inheritance, capital payment, and corporate taxes for any municipality, compare tax burdens, and find cheapest locations without an API key.
README
Swiss Tax Calculator — MCP Server & Python API
Licensed under MIT by Aiyatullah Saiyed.
A dual-purpose Swiss tax toolkit: use it as a Python API client in your applications, or run it as an MCP server for AI assistants like Claude. Wraps the official ESTV Swiss Federal Tax Calculator — no API key or authentication required.
Covers income, wealth, lump-sum capital, inheritance/gift, and corporate taxes for every Swiss municipality across tax years 2010–2026.
Features
| Category | Tools |
|---|---|
| Household tax | find_location, calculate_tax, list_deductions, calculate_tax_from_taxable_amounts |
| Compare places | compare_locations, find_cheapest_nearby |
| Tax planning | plan_capital_withdrawals, deduction_value |
| Other taxes | calculate_capital_payment_tax, calculate_inheritance_tax, calculate_company_tax |
| Reference | explain_tax_brackets, get_tax_years |
Quick Start
Install
# Clone the repository
git clone https://github.com/aiyatullah/swiss-tax-calculator-mcp.git
cd swiss-tax-calculator-mcp
# Install dependencies
uv sync
Usage as Python API
Import EstvClient directly in your Python code to query the ESTV tax calculator programmatically.
Basic Example — Find a Location
from estv_mcp import EstvClient
client = EstvClient()
# Search for a municipality by name or postal code
locations = client.search_location("Zug", tax_year=2026)
for loc in locations[:3]:
print(f"{loc['City']} ({loc['Canton']}) — ID: {loc['TaxLocationID']}")
Calculate Income Tax
from estv_mcp.api import EstvClient, RELATIONSHIP, CONFESSION, INCOME_TYPE
client = EstvClient()
# Step 1: Find the tax location
locations = client.search_location("8001", tax_year=2026)
location_id = locations[0]["TaxLocationID"]
# Step 2: Get the tax budget (deduction sheet)
budget = client.tax_budget({
"SimKey": None,
"TaxYear": 2026,
"TaxLocationID": location_id,
"Relationship": RELATIONSHIP["single"],
"Confession1": CONFESSION["none"],
"Confession2": 0,
"Children": [],
"Age1": 35,
"RevenueType1": INCOME_TYPE["employed"],
"Revenue1": 120_000,
"Age2": 0,
"RevenueType2": 0,
"Revenue2": 0,
"Fortune": 50_000,
})
# Step 3: Calculate detailed taxes
result = client.detailed_taxes({
"SimKey": None,
"TaxYear": 2026,
"TaxLocationID": location_id,
"Relationship": RELATIONSHIP["single"],
"Confession1": CONFESSION["none"],
"Confession2": 0,
"Children": [],
"Age1": 35,
"RevenueType1": INCOME_TYPE["employed"],
"Revenue1": 120_000,
"Age2": 0,
"RevenueType2": 0,
"Revenue2": 0,
"Fortune": 0,
"Budget": budget,
})
print(f"Total tax: CHF {result['TotalTax']:,.0f}")
print(f"Federal: CHF {result.get('IncomeTaxFed', 0):,.0f}")
print(f"Cantonal: CHF {result.get('IncomeTaxCanton', 0):,.0f}")
print(f"Municipal: CHF {result.get('IncomeTaxCity', 0):,.0f}")
Compare Municipalities
from estv_mcp.api import EstvClient, RELATIONSHIP, CONFESSION, INCOME_TYPE, GROUP_CAPITALS
client = EstvClient()
# Compare tax burden across all 26 cantonal capitals
rows = client.many_simple_taxes({
"SimKey": None,
"TaxYear": 2026,
"TaxGroupID": GROUP_CAPITALS,
"Relationship": RELATIONSHIP["single"],
"Confession1": CONFESSION["none"],
"Confession2": 0,
"Children": [],
"Age1": 35,
"RevenueType1": INCOME_TYPE["employed"],
"Revenue1": 100_000,
"Age2": 0,
"RevenueType2": 0,
"Revenue2": 0,
"Fortune": 0,
})
ranked = sorted(rows, key=lambda r: r["TotalTax"])
print("Top 5 cheapest cantonal capitals:")
for r in ranked[:5]:
loc = r["Location"]
print(f" {loc['City']} ({loc['Canton']}): CHF {r['TotalTax']:,.0f}")
Inheritance Tax
from estv_mcp.api import EstvClient, BENEFICIARY, GROUP_CAPITALS
client = EstvClient()
# Inheritance tax for a sibling across cantonal capitals
rows = client.many_inheritance_taxes({
"SimKey": None,
"TaxYear": 2026,
"TaxGroupID": GROUP_CAPITALS,
"OnlyGroupID": BENEFICIARY["sibling"][0],
"OnlyPersonID": BENEFICIARY["sibling"][1],
"Donation": False,
"Amount": 500_000,
})
ranked = sorted(rows, key=lambda r: r["TaxTotal"])
print(f"Cheapest: {ranked[0]['Location']['City']} — CHF {ranked[0]['TaxTotal']:,.0f}")
print(f"Most expensive: {ranked[-1]['Location']['City']} — CHF {ranked[-1]['TaxTotal']:,.0f}")
Available API Operations
| Method | Description |
|---|---|
search_location(query, tax_year) |
Find municipalities by name or postal code |
search_location_geo(lat, lon, radius_km, tax_year) |
Find municipalities within a radius |
tax_budget(payload) |
Get the deduction/budget sheet for a household |
detailed_taxes(payload) |
Full tax calculation with breakdown |
simple_taxes(payload) |
Tax from pre-computed taxable amounts |
many_simple_taxes(payload) |
Compare tax across many municipalities |
many_capital_taxes(payload) |
Lump-sum capital withdrawal tax across municipalities |
many_inheritance_taxes(payload) |
Inheritance/gift tax across municipalities |
many_legal_entity_taxes(payload) |
Corporate tax across municipalities |
export_tax_scales(tax_year, tax_group_id) |
Raw tax bracket tables |
tax_year_range(calculator) |
Supported year range per calculator |
tax_version() |
Current ESTV data version |
Enum Reference
from estv_mcp.api import (
RELATIONSHIP, # single=1, married=2, concubinage=3, registered_partnership=4
CONFESSION, # reformed=1, roman_catholic=2, christ_catholic=3, none=4, other=5
INCOME_TYPE, # employed=1, self_employed=2, pensioner=3, other=4
GENDER, # male=1, female=2
LANGUAGE, # de=1, fr=2, it=3, en=4
CANTON_GROUP, # AG=1 ... ZH=26
GROUP_CAPITALS, # 88 — all 26 cantonal capitals
GROUP_SWITZERLAND, # 99 — every municipality (~2100)
CALCULATOR, # income_wealth=1, capital_payment=2, legal_entity=3, inheritance=5
BENEFICIARY, # spouse, child, sibling, unrelated, etc. → (GroupID, PersonID)
)
Caching
Responses are cached under ~/.cache/estv-mcp for one week. Control via environment variables:
| Variable | Effect |
|---|---|
ESTV_MCP_NO_CACHE=1 |
Disable caching entirely |
ESTV_MCP_CACHE_DIR=/path |
Custom cache directory |
ESTV_MCP_CACHE_TTL=3600 |
Cache lifetime in seconds |
Usage as MCP Server
Run this project as an MCP server to give AI assistants (Claude, etc.) access to all Swiss tax tools.
Register with Claude Code
# Project-local (this repo only)
claude mcp add estv-tax -- uv run --directory "$PWD" estv-mcp
# User-wide (available in every project)
claude mcp add -s user estv-tax -- uv run --directory "$PWD" estv-mcp
MCP Client Config (JSON)
For any MCP-compatible client, add to your config:
{
"mcpServers": {
"estv-tax": {
"command": "uv",
"args": ["run", "--directory", "/path/to/swiss-tax-calculator-mcp", "estv-mcp"]
}
}
}
What the AI Can Do
Once connected, the AI assistant can:
- "How much tax would I pay in Zurich on CHF 150,000?" — calls
find_location+calculate_tax - "Compare Zug vs Schwyz for a married couple with 2 kids" — calls
compare_locations - "Where's the cheapest place within 30km of Zurich HB?" — calls
find_cheapest_nearby - "Should I split my pillar 2 withdrawal over multiple years?" — calls
plan_capital_withdrawals - "What's the inheritance tax for a sibling in Geneva?" — calls
calculate_inheritance_tax - "How much would my company pay in Zug vs Lucerne?" — calls
calculate_company_tax
Development
uv sync # Install dependencies (including dev)
uv run pytest -q # Run tests (hits live ESTV API)
uv run python scripts/stdio_smoke.py # MCP stdio handshake test
uv run ruff check . && uv run ruff format --check . # Lint & format check
Golden Values
tests/golden.json pins known tax figures for closed tax years. The weekly CI run re-checks them against the live API to detect upstream changes.
uv run python scripts/update_golden.py # Update golden values (prints diffs)
CI
Tests run on Python 3.11/3.12/3.13 on every push, plus a weekly schedule to catch ESTV data changes.
Important Notes
- All amounts are CHF per year
income_type='employed'means gross salary — social contributions (AHV/IV/EO, ALV, NBU, BVG) are derived automatically- Override deductions via
deductions={id: value}using IDs fromlist_deductions(e.g.PRAEMIEN3Afor pillar 3a,SCHULDZINSENfor mortgage interest) compare_locations(scope='switzerland')covers ~2,100 municipalities; results are summarized to stay readable- Church tax only applies when
confessionis notnone - Figures are from the official ESTV model — not a binding tax assessment
License
MIT License — see LICENSE.
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.