Pokémon MCP Server
Enables AI models to access comprehensive Pokémon data from PokéAPI and simulate battles between any two Pokémon with realistic mechanics including type effectiveness, stat-based damage calculations, and status effects.
README
Pokémon MCP Server — Data Resource + Battle Simulation Tool
Overview
This project implements an MCP (Model Context Protocol) server that provides AI models with access to two key capabilities:
1- Pokémon Data Resource – a resource that exposes comprehensive Pokémon data from the public PokéAPI ( https://pokeapi.co/ ).
2- Battle Simulation Tool – a tool that simulates battles between any two Pokémon, including type effectiveness, stats-based damage, turn order, and basic status effects.
This server acts as a bridge between AI and the Pokémon world, enabling LLMs to both retrieve knowledge and interactively simulate battles.
Part 1: Pokémon Data Resource
Implementation
-
Connects to the public PokéAPI ( https://pokeapi.co/ )
-
Exposes comprehensive Pokémon information including:
-
Base stats: HP, Attack, Defense, Special Attack, Special Defense, Speed
-
Types (e.g., Fire, Water, Grass)
-
Abilities
-
Available moves and their effects (power, accuracy, type, effect text)
-
Evolution information
MCP Resource
- Resource: pokemon://{name}
- Returns JSON including stats, types, abilities, moves (with effects), and evolution chain.
- Implements MCP resource design patterns to make this data accessible to LLMs.
Deliverables
- Code for the MCP server with the Pokémon data resource.
- Documentation (this README) describing how the resource exposes data.
- Example queries (examples/llm_examples.md).
Part 2: Battle Simulation Tool
Implementation
-
Tool: simulate_battle(pokemon_a, pokemon_b, max_turns=100, seed=None)
-
Simulates a battle between any two Pokémon using:
-
Type effectiveness calculations (e.g., Water > Fire)
-
Damage calculations based on stats and move power
-
Turn order based on Speed stat
Status effects:
- Paralysis – chance to skip a turn
- Burn – recurring HP loss
- Poison – recurring HP loss
- Detailed battle logs showing each turn’s actions and outcomes
- Winner determination (first Pokémon to faint, or higher HP after max turns)
MCP Tool
- Exposed via MCP as a callable tool: simulate_battle
- Returns JSON object, e.g.:
{
"winner": "blastoise",
"log": [
"--- Turn 1 ---",
"charizard used fire-punch → blastoise lost 14 HP!",
"blastoise is now affected by burn!",
...
]
}
Deliverables
- Code for the battle simulation tool following MCP tool specification
- Example usage in examples/llm_examples.md
Project Packaging
The submission includes a ZIP file containing:
- All code (pkmon_core/server.py, pkmon_core/battle.py)
- Supporting files (requirements.txt, README.md, examples/)
- A test script (test_part1.py)
- Clear instructions for setup and usage
Installation & Setup
Requirements
- Python 3.10+
- Virtual environment recommended
Setup
git clone https://github.com/TaifAlsadoon/pokemon-mcp-server.git
cd pokemon-mcp-server
python -m venv .venv
source .venv/bin/activate
pip install -r requirements.txt
Running & Testing
Run the MCP Server
python -m pkmon_core.server
Note: The server runs in stdio mode and will appear idle, waiting for an MCP client. Stop with Ctrl+C.
Quick Python Tests
python - << 'PY'
from pkmon_core.server import get_pokemon, simulate_battle
print("Resource sample (pikachu) ->")
print(get_pokemon("pikachu")[:200], "...\n")
print("Battle sample ->")
res = simulate_battle("charizard", "blastoise", seed=42, max_turns=20)
print("Winner:", res["winner"])
print("\n".join(res["log"][:10]))
PY
Expected output:
- JSON data for Pikachu (types, stats, moves, evolution chain)
- Battle log with turn-by-turn actions and a winner (e.g., Blastoise)
Examples for LLM Usage
See examples/llm_examples.md for prompt examples, such as:
- Summarizing a Pokémon’s stats, moves, and evolution
- Simulating a battle and explaining why the winner won
See examples/llm_examples.md for prompt examples.
Notes
- Simplified mechanics: ignores PP, items, weather, etc.
- Focused on clarity and educational battle simulation
- Easily extensible to add more mechanics
推荐服务器
Baidu Map
百度地图核心API现已全面兼容MCP协议,是国内首家兼容MCP协议的地图服务商。
Playwright MCP Server
一个模型上下文协议服务器,它使大型语言模型能够通过结构化的可访问性快照与网页进行交互,而无需视觉模型或屏幕截图。
Audiense Insights MCP Server
通过模型上下文协议启用与 Audiense Insights 账户的交互,从而促进营销洞察和受众数据的提取和分析,包括人口统计信息、行为和影响者互动。
Magic Component Platform (MCP)
一个由人工智能驱动的工具,可以从自然语言描述生成现代化的用户界面组件,并与流行的集成开发环境(IDE)集成,从而简化用户界面开发流程。
VeyraX
一个单一的 MCP 工具,连接你所有喜爱的工具:Gmail、日历以及其他 40 多个工具。
Kagi MCP Server
一个 MCP 服务器,集成了 Kagi 搜索功能和 Claude AI,使 Claude 能够在回答需要最新信息的问题时执行实时网络搜索。
graphlit-mcp-server
模型上下文协议 (MCP) 服务器实现了 MCP 客户端与 Graphlit 服务之间的集成。 除了网络爬取之外,还可以将任何内容(从 Slack 到 Gmail 再到播客订阅源)导入到 Graphlit 项目中,然后从 MCP 客户端检索相关内容。
mcp-server-qdrant
这个仓库展示了如何为向量搜索引擎 Qdrant 创建一个 MCP (Managed Control Plane) 服务器的示例。
e2b-mcp-server
使用 MCP 通过 e2b 运行代码。
Neon MCP Server
用于与 Neon 管理 API 和数据库交互的 MCP 服务器