Roblox MCP Server

Roblox MCP — MCP Server for Roblox Studio | AI Game Development with Claude, Codex, Cursor & Gemini

Weppy Roblox MCP is an MCP server that lets AI coding agents control a live Roblox Studio session — create and edit scripts, instances, terrain, lighting, assets, audio, and animations via natural language.

21 consolidated tools · 140+ actions · Bidirectional sync · Automated playtest · Multi-place support

English | 한국어 | 日本語 | Español | Português | Bahasa Indonesia

Why Weppy Roblox MCP?

AI coding agents like Claude, Codex, and Gemini are powerful — but they cannot see or modify anything inside Roblox Studio. The DataModel, scripts, terrain, and lighting are all invisible to external tools. Without a bridge, AI can only generate code snippets that you must paste manually.

Weppy Roblox MCP provides a real-time bridge between AI agents and the Roblox Studio DataModel. It exposes the full Roblox API surface as executable MCP tools, so AI can directly create, read, modify, and delete instances, scripts, properties, terrain, and more — inside a live Studio session.

This is not just code generation. It is executable, production-oriented actions running directly in Studio.

Quick Install (5 minutes)

Step 1 — Install the Roblox Studio plugin (bridge between Studio and AI): Plugin Installation Guide

Step 2 — Register the MCP server with your AI app:

npx -y @weppy/roblox-mcp

Any MCP-compatible AI client works. The server command is npx -y @weppy/roblox-mcp.

Compatibility

Requirements: Node.js 18+, Roblox Studio, Windows 10+ or macOS 12+

What It Does

1) MCP Tools: Direct execution in Studio from natural language

AI can directly handle scripts, instances, properties, terrain, lighting, assets, audio, and animation inside Studio.

• "Add particles + sound + cooldown when the player jumps."
• "Build a boss arena at map center and place collision-safe spawn points."
• "Change this module interface and update every dependent script."
• "Generate terrain with mountains and rivers, then place spawn points on flat areas."
• "Search the Creator Store for a sword model and insert it into StarterPack."

2) Sync: Keep full project context stable for AI

AI works from a synchronized local mirror, so multi-file updates stay consistent.

• Basic: one-way sync (Studio -> Local)
• Pro: bidirectional sync + per-type Direction/Apply Mode + history + multi-place

3) Playtest: Let AI run and verify tests automatically

AI can control Roblox Studio playtests directly. It can start and stop Play (F5) or Run (F8), inject test scripts, collect logs, and generate local reports automatically.

• "Start a Run-mode playtest and check whether the NPC reaches the target."
• "Write a test that verifies the SpawnLocation is above the ground and run it."
• "Validate that the script I just changed runs without errors in playtest."

4) Dashboard: Monitor AI work in real time

The MCP server provides a web dashboard where you can check connection status, tool execution history, sync state, and game change logs in real time.

• Server/Plugin/Agent connection status at a glance
• Compare every change the AI made via Before & After in Changelog
• Analyze workflow with tool execution history and statistics

5) Roblox Explorer: Browse Studio hierarchy in VSCode

View the full instance tree of your Roblox Studio place directly inside VSCode. Navigate services, open synced scripts and property files, and track sync status — all without switching to Studio.

• Class icons matching Studio for instant recognition
• Click to open synced scripts and property files
• Multi-plac