Flame MCP Server
Provides live, up-to-date documentation for the Flame game engine.
Flame MCP Server
A Model Context Protocol (MCP) server that provides comprehensive access to Flame game engine documentation for AI assistants like Claude Desktop and Amazon Q Developer.
🎯 What This Does

- Documentation Access: Provides searchable access to the complete Flame engine documentation
- Tutorial System: Offers step-by-step game development tutorials (Space Shooter, Platformer, Klondike)
- Local Caching: Stores documentation locally for fast, offline access
- MCP Integration: Works seamlessly with Claude Desktop and Amazon Q CLI
- Search Tools: Intelligent search across all documentation and tutorials
🚀 Quick Start
1. Build and Setup
# Clone and build the server
git clone <repository-url>
cd flame_mcp_server
./build_clean.sh
This will:
- Install Dart dependencies
- Build the MCP server executable
- Download and cache all Flame documentation (~146 files)
2. Configure Your MCP Client
Amazon Q Developer
Add to your MCP configuration:
{
"mcpServers": {
"flame-docs": {
"command": "/absolute/path/to/flame_mcp_server/build/flame_mcp_live"
}
}
}
Claude Desktop
Add to claude_desktop_config.json:
{
"mcpServers": {
"flame-docs": {
"command": "/absolute/path/to/flame_mcp_server/build/flame_mcp_live"
}
}
}
Config file locations:
- macOS:
~/Library/Application Support/Claude/claude_desktop_config.json - Windows:
%APPDATA%\\Claude\\claude_desktop_config.json - Linux:
~/.config/Claude/claude_desktop_config.json
3. Restart Your MCP Client
Restart Claude Desktop or Amazon Q CLI to load the Flame documentation server.
🛠️ Available Tools
search_documentation
Search through all Flame documentation for specific topics.
Example queries:
- "How do I implement collision detection in Flame?"
- "Search for component system examples"
- "Find information about sprite animations"
tutorial
Get complete step-by-step game development tutorials.
Available tutorials:
- Space Shooter: Complete 6-step tutorial for building a classic space shooter
- Platformer: 7-step tutorial for building a side-scrolling platformer (Ember Quest)
- Klondike: 5-step tutorial for building a solitaire card game
Example usage:
- "Show me how to build a space shooter game" → Complete tutorial with all steps
- "I want to create a platformer game" → Full platformer tutorial
- "List all available tutorials" → Overview of all tutorials
📊 Architecture
┌─────────────────┐ ┌──────────────────┐ ┌─────────────────┐
│ MCP Client │ │ Flame MCP │ │ Documentation │
│ (Claude/Amazon Q)│◄──►│ Server │◄──►│ Cache │
└─────────────────┘ └──────────────────┘ └─────────────────┘
│
▼
┌──────────────────┐
│ GitHub API │
│ (flame-engine/ │
│ flame/doc) │
└──────────────────┘
🔧 Configuration
GitHub Token (Recommended)
For better rate limits (5,000 vs 60 requests/hour):
-
Create a GitHub Personal Access Token:
- Go to GitHub Settings → Developer settings → Personal access tokens
- Generate new token with
public_reposcope
-
Set environment variable:
# macOS/Linux export GITHUB_TOKEN=your_token_here # Windows setx GITHUB_TOKEN "your_token_here"
📁 Project Structure
flame_mcp_server/
├── bin/
│ ├── flame_mcp_live.dart # Main MCP server
│ └── flame_sync_standalone.dart # Documentation sync utility
├── lib/src/
│ ├── flame_live_docs.dart # Documentation management
│ └── flame_mcp_live.dart # MCP protocol implementation
├── build/
│ └── flame_mcp_live # Compiled MCP server
├── flame_docs_cache/ # Cached documentation (146 files)
├── build_clean.sh # Build and setup script
└── README.md # This file
🎮 Example Interactions
Building a Space Shooter Game
You: "I want to build a space shooter game in Flame. Show me the complete tutorial."
Response: Complete 6-step tutorial including:
- Project setup and basic game structure
- Player controls and graphics
- Animations and visual effects
- Enemy spawning and movement
- Shooting mechanics
- Collision detection and scoring
Learning About Components
You: "How does the Flame component system work?"
Response: Detailed documentation about:
- Component lifecycle
- Component hierarchy
- Built-in components
- Creating custom components
- Component communication
🔄 Maintenance
Refresh Documentation Cache
# Update to latest Flame documentation
dart run bin/flame_sync_standalone.dart
Rebuild Server
# Clean rebuild with fresh documentation
./build_clean.sh
📋 Prerequisites
- Dart SDK: Version 3.2.0 or higher
- Internet Connection: Required for initial documentation sync
- MCP Client: Claude Desktop, Amazon Q CLI, or compatible client
🐛 Troubleshooting
Server Not Found
- Ensure you're using the absolute path to the executable in your MCP config
- Verify the executable exists:
ls -la build/flame_mcp_live - Check file permissions:
chmod +x build/flame_mcp_live
No Search Results
- Run
./build_clean.shto rebuild cache and server - Check cache exists:
ls flame_docs_cache/ - Verify cache has content:
find flame_docs_cache -name "*.md" | wc -l(should show ~146)
Rate Limit Issues
- Set up a GitHub personal access token (see Configuration section)
- Check rate limit status in sync logs
📈 Performance
- Documentation Files: 146 Markdown files
- Cache Size: ~3 MB
- Sync Time: 30-60 seconds (network dependent)
- Memory Usage: <50 MB when running
- Startup Time: <2 seconds
🤝 Contributing
- Fork the repository
- Create a feature branch
- Make your changes
- Test with
./build_clean.sh - Submit a pull request
📄 License
This project is licensed under the MIT License.
🔗 Related Links
Ready to start building games with Flame? Run ./build_clean.sh and add the server to your MCP client! 🎮
Serveurs connexes
Alpha Vantage MCP Server
sponsorAccess financial market data: realtime & historical stock, ETF, options, forex, crypto, commodities, fundamentals, technical indicators, & more
Remote MCP Server (Authless)
An example of a remote MCP server deployable on Cloudflare Workers without authentication.
Proteus Workflow Engine
A modern, extensible multi-agent workflow engine with real-time monitoring and a web visualization interface.
VSCode MCP
Enables AI agents and assistants to interact with VSCode through the Model Context Protocol.
IdeaJarvis
IdeaJarvis is an idea workspace for product builders. Use AI to structure brainstorming into detailed PRDs, conduct comprehensive market research, build prototypes, and gather real community feedback—turning "what if" into "ready to launch.
Unified MCP & A2A Server
A Google Apps Script server that unifies Model Context Protocol (MCP) and Agent2Agent (A2A) for Google Workspace users.
weibaohui/kom
Provides multi-cluster Kubernetes management and operations using MCP, It can be integrated as an SDK into your own project and includes nearly 50 built-in tools covering common DevOps and development scenarios. Supports both standard and CRD resources.
SeedDream 3.0
Generate images using Bytedance's SeedDream 3.0 model via the FAL AI platform.
Honeybadger
Interact with the Honeybadger API for error monitoring and reporting using LLMs.
MCP Spine
Context Minifier & State Guard — Local-first MCP middleware proxy that reduces token waste by 61%, prevents context rot, and adds security hardening
Aptos NPM MCP
A MCP server for interacting with Aptos NPM packages.