PokeMCP

chiuwa/PokeMCP

3.2

If you are the rightful owner of PokeMCP and would like to certify it and/or have it hosted online, please leave a comment on the right or send an email to henry@mcphub.com.

The PokeAPI MCP Server is a Model Context Protocol server that interacts with the PokeAPI to provide tools for querying Pokémon and item information, designed for easy consumption by AI Agents.

Tools
7
Resources
0
Prompts
0

PokeAPI MCP Server

This project implements a Model Context Protocol (MCP) server using FastMCP. The server interacts with the PokeAPI to provide tools for querying Pokémon and item information. This server is designed to be easily consumable by AI Agents.

image image

Features

The server exposes the following tools:

  • get_pokemon_details(pokemon_name_or_id: str): Fetches comprehensive details for a specific Pokémon.
  • get_pokemon_types(pokemon_name_or_id: str): Fetches the elemental types of a specific Pokémon.
  • get_pokemon_color(pokemon_name_or_id: str): Fetches the Pokedex color of a specific Pokémon.
  • get_pokemon_shape(pokemon_name_or_id: str): Fetches the Pokedex shape category of a specific Pokémon.
  • list_all_pokemon_names(limit: int = 2000, offset: int = 0): Lists English names of Pokémon, supporting pagination.
  • get_item_details(item_name_or_id: str): Fetches detailed information about a specific in-game item.
  • list_all_items(limit: int = 100, offset: int = 0): Lists in-game items, supporting pagination.

Refer to the SERVER_INSTRUCTIONS variable and individual tool docstrings in server.py for more detailed information on each tool's parameters and return values.

Setup

  1. Prerequisites:

    • Python (Python 3.10 or newer recommended for httpx async features)
    • uv (Recommended: an extremely fast Python package installer and resolver. )
    • Alternatively, pip and virtualenv (or venv if not using uv)
  2. Create and activate a virtual environment:

    Recommended (using uv):

    # Create a virtual environment (e.g., in .venv)
    uv venv
    # Activate it
    source .venv/bin/activate  # On Windows: .venv\Scripts\activate
    

    Alternative (using venv):

    python -m venv venv
    source venv/bin/activate  # On Windows: venv\Scripts\activate
    
  3. Install dependencies:

    Recommended (using uv):

    uv pip install -r requirements.txt
    

    Alternative (using pip):

    pip install -r requirements.txt
    

Running the Server

To start the MCP server:

python server.py

By default, the server will run on http://0.0.0.0:7796 using Server-Sent Events (SSE) transport. You should see logging output indicating the server has started and is listening for connections.

Ensure the server is running before starting the client. The client will attempt to connect to the server URL specified within it (defaulting to http://localhost:7796/sse).


PokeAPI MCP 伺服器

本專案使用 FastMCP 實作了一個模型上下文協定 (MCP) 伺服器。該伺服器與 PokeAPI 互動,提供查詢寶可夢 (Pokémon) 和道具資訊的工具。此伺服器旨在方便 AI Agent 使用。

主要功能

伺服器提供以下工具:

  • get_pokemon_details(pokemon_name_or_id: str): 獲取特定寶可夢的詳細資訊。
  • get_pokemon_types(pokemon_name_or_id: str): 獲取特定寶可夢的屬性。
  • get_pokemon_color(pokemon_name_or_id: str): 獲取特定寶可夢的圖鑑顏色。
  • get_pokemon_shape(pokemon_name_or_id: str): 獲取特定寶可夢的圖鑑分類外形。
  • list_all_pokemon_names(limit: int = 2000, offset: int = 0): 列出寶可夢的英文名稱,支援分頁。
  • get_item_details(item_name_or_id: str): 獲取特定遊戲內道具的詳細資訊。
  • list_all_items(limit: int = 100, offset: int = 0): 列出遊戲內的道具,支援分頁。

關於每個工具的參數和回傳值的更詳細資訊,請參考 server.py 中的 SERVER_INSTRUCTIONS 變數和各個工具的 docstrings。

設定步驟

  1. 先決條件:

    • Python (建議使用 Python 3.10 或更新版本以獲得 httpx 的非同步特性支援)
    • uv (推薦:一個極速的 Python 套件安裝器和解析器。)
    • 或者,如果您不使用 uv,則需要 pipvirtualenv (或 venv)
  2. 建立並啟用虛擬環境:

    推薦 (使用 uv):

    # 建立虛擬環境 (例如在 .venv 資料夾中)
    uv venv
    # 啟用虛擬環境
    source .venv/bin/activate  # Windows 環境: .venv\Scripts\activate
    

    替代方案 (使用 venv):

    python -m venv venv
    source venv/bin/activate  # Windows 環境: venv\Scripts\activate
    
  3. 安裝依賴套件:

    推薦 (使用 uv):

    uv pip install -r requirements.txt
    

    替代方案 (使用 pip):

    pip install -r requirements.txt
    

運行伺服器

啟動 MCP 伺服器:

python server.py

預設情況下,伺服器將使用 Server-Sent Events (SSE) 傳輸方式在 http://0.0.0.0:7796 上運行。您應該能看到日誌輸出,顯示伺服器已啟動並正在監聽連線。

在啟動客戶端之前,請確保伺服器正在運行。客戶端將嘗試連接到其內部指定的伺服器 URL (預設為 http://localhost:7796/sse)。