home / mcp / parking mcp server

Parking MCP Server

Provides nationwide parking data via MCP using Kakao Local, Seoul Open Data, and Gyeonggi Data Dream APIs.

Installation

Add the following to your MCP client configuration file.

Configuration

{
  "mcpServers": {
    "acorn025-parking-mcp2": {
      "command": "python",
      "args": [
        "-m",
        "src.server"
      ],
      "env": {
        "PYTHONPATH": "src",
        "KAKAO_REST_API_KEY": "YOUR_KAKAO_API_KEY",
        "SEOUL_DATA_API_KEY": "YOUR SEOUL DATA API KEY",
        "GYEONGGI_DATA_API_KEY": "YOUR GYEONGGI DATA API KEY"
      }
    }
  }
}

You run this MCP server to query nationwide parking lot information and provide real-time data for specific regions. It consolidates data sources from a national search API and regional open data feeds to give you up-to-date parking availability, operating hours, and fees in an easy-to-consume format for client applications.

How to use

To use this MCP server, start the server locally and connect a client configured to the MCP protocol. The server exposes functions to search nearby parking and to fetch detailed parking information. Your client can request data for any location, and you’ll receive structured results that include basic parking details, availability where available, and regional information.

Typical usage pattern: - Start the MCP server on your machine. - Configure your MCP client to connect to the local server using the provided MCP command configuration. - Call search_nearby_parking with latitude, longitude, and an optional radius to discover nearby parking facilities. - Use get_parking_info to retrieve detailed information for a chosen parking place by name or address. - Rely on region-specific data if you query 서울 (Seoul) or 경기 (Gyeonggi) areas for enhanced details.

How to install

Prerequisites you need before installing: - Python 3.8 or higher - Access to a command shell (PowerShell on Windows or Command Prompt) - Internet access to install dependencies and fetch data sources.

# 1) Create and activate a virtual environment
python -m venv venv

# Windows (PowerShell)
.venv\Scripts\Activate.ps1

# Windows (PowerShell execution policy note)
Set-ExecutionPolicy -ExecutionPolicy RemoteSigned -Scope CurrentUser

# Alternative (cmd)
venv\Scripts\activate.bat

# 2) Install Python dependencies
pip install --upgrade pip
pip install -r requirements.txt

# 3) Create the environment variables file
copy .env.example .env

After creating the .env file, open it and insert your API keys for Kakao Local, Seoul Open Data, and Gyeonggi Data Dream as described in the keys section.

Additional information

Configuration and runtime notes: - Start the server module directly so it runs under the MCP framework. The runtime command is: - python -m src.server - The server expects environment variables to be present, including keys for the Kakao Local API and regional data sources. Ensure these keys are populated in .env before starting. - When you run the server, you’ll typically set PYTHONPATH to include the source directory so the module can be discovered by Python.

Security and maintenance tips: - Keep your API keys secret and avoid committing .env into version control. - Update dependencies regularly by refreshing requirements.txt and rebuilding the virtual environment as needed. - Monitor access to the MCP endpoint and rotate keys if you suspect compromise.

Available tools

search_nearby_parking

Searches for nearby parking facilities using the national Kakao Local API, with distance-based results and regional information added automatically for Seoul, Gyeonggi, or other areas.

get_parking_info

Retrieves detailed information for a specific parking place by name or address, including regional data when available.