Cardossier Market API MCP server

CarDossier Poland Market API — MCP Server

mcp-name: io.github.Joyall-au/cardossier-market-api

An MCP (Model Context Protocol) server that gives AI assistants — Claude, Cursor, Windsurf, and any MCP-compatible agent — real-time access to Polish used car market data via the CarDossier Market API.

What It Does

This server exposes 5 tools that AI agents can call to answer questions like:

• "What is a 2019 Toyota Corolla worth in Poland right now?"
• "Has the price of diesel BMWs been rising or falling over the past 6 months?"
• "How quickly do used Volkswagen Golfs sell in Poland?"
• "Is a 2020 Audi A4 cheaper in Mazowieckie or Śląskie?"
• "How much more does an automatic gearbox add to the price of a 2018 Ford Focus?"

Available Tools

Data source: 1.4M+ active listings from Poland's leading car marketplaces, updated daily.

Vehicle names are resolved automatically — VW Golf, BMW 3 Series / 320d, Mercedes C-Class, Audi A4, XC60 all work; responses show exactly which marketplace models were aggregated. Failed calls are auto-refunded.

🚀 Try It Without a Key

The API serves 5 keyless demo calls per IP per day with full real data, so this server works the moment you install it:

pip install cardossier-mcp-server
cardossier-mcp

For more quota, register free — 50 credits via email OTP, no credit card — and set CARDOSSIERAPIKEY.

Prerequisites

• Python 3.9+
• Optional: a CarDossier API key — register for free to get 50 free credits (no credit card required), or purchase a paid package starting at $49 for 5,000 credits

Installation

# From PyPI (recommended)
pip install cardossier-mcp-server

# Or from source
git clone https://github.com/Joyall-au/cardossier-mcp-server.git
cd cardossier-mcp-server
pip install mcp requests

Usage

1. (Optional) Set your API key

export CARDOSSIER_API_KEY="your_api_key_here"   # omit to use the keyless demo tier

2. Run the server

cardossier-mcp
# or from a source checkout: python mcp_server.py

3. Connect to Claude Desktop

Add the following to your Claude Desktop config file:

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

{
  "mcpServers": {
    "cardossier": {
      "command": "cardossier-mcp",
      "env": {
        "CARDOSSIER_API_KEY": "your_api_key_here"
      }
    }
  }
}

4. Connect to Cursor or Windsurf

Add to your MCP settings:

{
  "cardossier": {
    "command": "cardossier-mcp",
    "env": {
      "CARDOSSIER_API_KEY": "your_api_key_here"
    }
  }
}

Example Interactions

Once connected, you can ask your AI assistant:

"What's the current market value of a 2020 BMW 3 Series diesel in Poland?"

The agent will call getmarketvaluation with make=BMW, model=3 Series, year=2020, fuel_type=diesel and return the statistical distribution of prices.

"Show me how Toyota Corolla prices have changed over the last year"

Shortened. The full README is on GitHub.