Appearance
AI Assistants
Practical tips for using AI assistants with Metaplay.
Appearance
Practical tips for using AI assistants with Metaplay.
Metaplay works exceptionally well with AI coding assistants. The SDK ships as source code with comprehensive documentation and full sample projects, giving AI tools full context to understand patterns, conventions, and implementation details.
Basically, Metaplay's core architecture and language choices are a great fit to AI assistants!
LLMs Have Their Limits
While AI tool are the best way to quickly explore the Metaplay SDK, LLMs do not know everything about Metaplay or your needs and are prone to simple assumptions. For deep diving architecture questions or topic not covered by our docs, please contact our support or sales!
The easiest AI assistant to try is our Metaplay Docs chatbot, hosted by OpenAI.

It can help you:
You can get creative and even ask it to search the web for common game design features and how to approach making them with Metaplay!
To go a step further, you can use the Metaplay Docs model context protocol (MCP) server to make our documentations, SDK source, and samples available to your own AI coding assistants. All major IDEs and agentic coding tools, such as Cursor, Claude Code, Codex, Windsurf, Visual Studio Code, and others support MCP servers.
By adding the MCP server to your AI coding assistant, it suddenly has access to over 10x more Metaplay content than by relying on web searches! Especially the very latest large coding models benefit greatly from this.
To set up the MCP server in your tool of choice, use this URL:
https://llm-docs.platform.metaplay.dev/mcpClaude is an AI assistant by Anthropic that focuses on business use cases. Their web, mobile, and desktop versions all support MCP servers, making it ideal for more than just web browsing.
Under Settings -> Connectors, add a new connector with the URL https://llm-docs.platform.metaplay.dev/mcp. You can name it "Metaplay Docs".


You can now ask Claude about Metaplay! This is especially useful if you give Claude access to your design documents and code, so it can help you plan new features.
Claude Code is an agentic coding assistant by Anthropic. We feel their Opus model is particularly good at programming games with Metaplay.
You can add the MCP server with the following command:
claude mcp add --transport http --scope project metaplay-docs https://llm-docs.platform.metaplay.dev/mcpThe command is a helper to create the config file for the MCP that you can commit to source control and share with your team:
{
"mcpServers": {
"metaplay-docs": {
"type": "http",
"url": "https://llm-docs.platform.metaplay.dev/mcp"
}
}
}
Your Claude Code programming agents will now have access to Metaplay documentation, SDK source, and sample projects!
OpenCode is an open source AI coding agent. It’s available as a terminal-based interface, desktop app, or IDE extension. You can use it with a lot of free models!
You can add our MCP server to the OpenCode settings file in your repo root:
{
"$schema": "https://opencode.ai/config.json",
"mcp": {
"metaplay-docs": {
"type": "remote",
"url": "https://llm-docs.platform.metaplay.dev/mcp",
"enabled": true
}
}
}
Your OpenCode agents will now have access to Metaplay documentation, SDK source, and sample projects!