Lol_Data_MCP_Server: Instructions file for Claude Code

CLAUDE.md

This file provides guidance to Claude Code (claude.ai/code) when working with code in this repository.

🎯 PROJECT OVERVIEW

LoL Data MCP Server - A comprehensive Model Context Protocol server that provides real-time access to League of Legends game data including champions, items, abilities, game mechanics, and patch information.

Project Context

This is part of Project Taric, which contains 3 separate projects:

• LoL_Data_MCP_Server (this project): MCP server for League of Legends data
• Lol_Sim_Env: LoL simulation environment for AI training
• Taric_AI_Agent: AI agent that plays Taric champion

Each project has its own virtual environment and focuses on different aspects of the LoL AI ecosystem.

🔄 DEVELOPMENT ENVIRONMENT SETUP

CRITICAL: Activate the virtual environment for Python operations:

# Navigate to project directory 
cd LoL_Data_MCP_Server

# Activate virtual environment (Windows)
.\venv\Scripts\Activate.ps1

# Verify activation - should see (venv) in terminal prompt

When to use venv:

• ✅ Running tests: pytest
• ✅ Running MCP server: python -m src.mcp_server.stdio_server
• ✅ Code quality tools: black ., mypy ., flake8 ., isort .
• ✅ Installing dependencies: pip install -r requirements.txt

When venv is NOT needed:

• ❌ File editing/reading
• ❌ Git operations: git add, git commit, git push
• ❌ Basic file operations

🚨 CRITICAL QUESTIONS RULE

When you have REALLY IMPORTANT questions about the project or during task execution:

• YOU CAN ASK QUESTIONS even if it stops the generation progress
• Important questions include:
  • Architecture decisions that affect multiple components
  • Unclear requirements that could lead to wrong implementation
  • Breaking changes that might affect existing functionality
  • Technical approach questions when multiple valid options exist
  • Data model or API design questions
  • Integration concerns with existing systems
• When to ask: If the question is critical for correct implementation
• How to ask: Stop generation, ask the question clearly, wait for clarification
• Priority: Getting the right answer is more important than continuous generation