A gameboy emulator for LLM's
MCP Gameboy connects MCP Gameboy to AI assistants that speak the Model Context Protocol. A gameboy emulator for LLM's.
You can... - control the gameboy emulator using the @modelcontextprotocol/inspector - control the gameboy emulator (and upload ROMs) using a web-interface at http://localhost:3001/emulator - install the gameboy emulator in your favorite MCP-Client
A Model Context Protocol (MCP) server for serverboy, allowing LLMs to interact with a GameBoy emulator. Your LLM can... - Load ROMS - Press Keys - Look at the Gameboy Screen - skip frames
Compare MCP servers by category, or browse hosted MCP servers if you prefer an endpoint over a local process.
Check the linked publisher documentation and your connected client's tool list. Names and availability can vary with the server version and account permissions.
| Tool | What it does |
|---|---|
| press_up | Press the UP button on the GameBoy |
| press_down | Press the DOWN button on the GameBoy |
| press_left | Press the LEFT button on the GameBoy |
| press_right | Press the RIGHT button on the GameBoy |
| press_a | Press the A button on the GameBoy |
| press_b | Press the B button on the GameBoy |
| press_start | Press the START button on the GameBoy |
| press_select | Press the SELECT button on the GameBoy |
| load_rom | Load a GameBoy ROM file |
| get_screen | Get the current GameBoy screen |
{
"mcpServers": {
"mcp-gameboy": {
"command": "node",
"args": [
"/path/to/mcp-gameboy/dist/index.js"
],
"disabled": false,
"autoApprove": []
}
}
}From the project documentation. Replace placeholders and example paths before connecting.
Read, search and edit files inside allowed directories with the Model Context Protocol reference Filesystem server.
Retrieve version-aware library documentation and code examples with Context7 MCP, using resolve-library-id and query-docs.
This MCP server provides both general command management tools and **dynamic command-specific tools**:
AI coding agents are powerful -- but without structure, they wander. They lose track of the plan. They skip tests. They write code that works but
Read-only DAGP governance rules, agent quickstart and experimental society guides.
目前博查的搜索效果是国内最接近Bing Search API的搜索引擎，由于Bing Search API数据会出海（无国内Region）、价格昂贵（15美元/千次）且不提供文本摘要（只有50-100字的snippet），国内很多企业客户都已经从Bing切换至博查。