League-of-Legends
Unified MCP server for all League of Legends Riot API endpoints.
claude mcp add --transport stdio kostadindev-league-of-legends-mcp docker run -i kostadindev/league-mcp:latest \ --env RIOT_API_KEY="your Riot Games API key (obtain from https://developer.riotgames.com/)"
How to use
This MCP server exposes Riot Games APIs, LoL static resources (champion data, items, queues, maps, etc.), and workflow prompts to support building prompts and tools for game-related tasks. It also bundles a client with a ReAct-based agent UI to demonstrate tool/resource/prompt calls. You can run it inside Docker and connect to the provided endpoints to query Riot data, inspect static resources, and experiment with workflow prompts such as player stats analysis, champion analysis, and team composition suggestions. When using the docker option, you will need to supply your Riot API key via the RIOT_API_KEY environment variable before starting the container. The server will expose an API surface that the MCP client can call to retrieve data and run prompts through the ReAct agent UI.
How to install
Prerequisites:
- Docker installed on your machine
- Riot Games API key (create at https://developer.riotgames.com/)
-
Pull and run the MCP server using Docker: docker run -i -e RIOT_API_KEY=your_api_key_here kostadindev/league-mcp:latest
Optional: expose web UI or endpoints as needed by mapping ports.
-
If you prefer to use a docker-compose setup: version: '3.8' services: league-mcp: image: kostadindev/league-mcp:latest environment: - RIOT_API_KEY=${RIOT_API_KEY} command: ["league-mcp"] ports: - "8000:8000" restart: unless-stopped
-
Verify the container is running and accessible. Use your browser or API clients to connect to the exposed endpoints as documented by the project (e.g., the MCP interface and any UI provided).
Note: If you intend to run the client and server together locally, you may need to follow additional project-specific steps to install Python dependencies or use uv for faster setups, as described in the repository’s guidance.
Additional notes
Tips and common issues:
- Ensure RIOT_API_KEY is valid and has the required permissions for the Riot APIs you intend to use.
- If you’re also using the client UI, you may need a Google Gemini/Gradio setup for the client portion; set GOOGLE_API_KEY if you plan to enable client features.
- When using Docker, you can pass different transports or ports via the container command if the MCP supports them (e.g., a separate transport like SSE for web integrations).
- If you encounter rate limits, consider staggering requests or using a Riot API key with proper rate limits.
- The MCP server exposes static resources (champions, items, etc.). Use the provided prompts to analyze or query these resources in the client UI or via API calls.
- Review the README for any environment variable you might need to customize (e.g., optional transports, UI settings, or additional API keys).
Related MCP Servers
ebook
A MCP server that supports mainstream eBook formats including EPUB, PDF and more. Simplify your eBook user experience with LLM.
paiml -agent-toolkit
Pragmatic AI Labs MCP Agent Toolkit - An MCP Server designed to make code with agents more deterministic
infobus
Model Context Protocol server enabling AI assistants to access transit information through standardized interfaces
mcp-gateway
MCP Gateway and Registry
mcp -email
一个基于 MCP (Model Context Protocol) 的邮件服务,支持 LLM 发送带附件的电子邮件及在指定目录中搜索文件。提供安全的 SMTP 传输、多收件人支持和附件模式匹配搜索功能,适用于 Gmail、Outlook、Yahoo、QQ 邮箱和网易 126 邮箱等主流邮箱服务。
MCPSecBench
MCPSecBench: A Systematic Security Benchmark and Playground for Testing Model Context Protocols