MCP Kakao Local
# MCP Kakao Local
The MCP connects to the [Kakao Local API](https://developers.kakao.com/docs/latest/ko/local/common) and Kakao Map.
카카오 로컬 API 및 카카오맵에 연결하는 MCP 서버.
## Prerequisites
Before you begin, ensure you have the following installed:
* **Python:** Version 3.13 or higher
* **uv:** You can find installation instructions [here](https://github.com/astral-sh/uv).
* **Kakao Developers:** You need API credentials from the [Kakao developers](https://developers.kakao.com/).
## Configuration
1. **Create a `.env` file:** Create a file in the project root.
2. **Add API Credentials:** Edit the `.env` file and add your Kakao REST API credentials.
```.env
REST_API_KEY="YOUR_REST_API_KEY_HERE"
```
Please verify the exact environment variable names required by checking `src/mcp_kakao_local/kakao_local_client.py`.
## Running the MCP
1. **Sync Dependencies:** Navigate to the project root directory in your terminal and run the following command. This will create a virtual environment (if one doesn't exist) and install all dependencies specified in `pyproject.toml`.
```bash
uv sync
```
2. **Run:**: You can run the MCP server using `uv`.
```bash
uv run src/mcp_kakao_local
```
For development,
```bash
source .venv/bin/activate
mcp dev src/mcp_kakao_local/server.py
```
TDQS
Scored across 4 tools
Each tool has a clearly distinct purpose: find_coordinates converts addresses to coordinates, get_place retrieves detailed place information, search_by_category filters by category codes, and search_by_keyword searches by text keywords. There is no overlap in functionality, making tool selection unambiguous for an agent.
The tools follow a consistent verb_noun pattern (e.g., find_coordinates, get_place, search_by_category, search_by_keyword) with clear, descriptive names. The minor deviation is that 'find' and 'get' are used as verbs instead of a single verb style, but this does not hinder readability or predictability.
With 4 tools, the server is well-scoped for a local search and mapping service. Each tool serves a specific, essential function (geocoding, place details, category search, keyword search), and there are no extraneous tools, making the count appropriate for the domain.
The toolset covers core functionalities for a local search API: geocoding, place details, and two search methods (category and keyword). A minor gap is the lack of route planning or navigation tools, but the provided tools allow agents to perform key tasks like finding and exploring places effectively.