snapgent-connector

Turn any AI chat into an agent for Roblox Studio and VS Code.

herdo@snapgent:~$ snapgentV3

  ███████╗███╗   ██╗ █████╗ ██████╗  ██████╗ ███████╗███╗   ██╗████████╗
  ██╔════╝████╗  ██║██╔══██╗██╔══██╗██╔════╝ ██╔════╝████╗  ██║╚══██╔══╝
  ███████╗██╔██╗ ██║███████║██████╔╝██║  ███╗█████╗  ██╔██╗ ██║   ██║
  ╚════██║██║╚██╗██║██╔══██║██╔═══╝ ██║   ██║██╔══╝  ██║╚██╗██║   ██║
  ███████║██║ ╚████║██║  ██║██║     ╚██████╔╝███████╗██║ ╚████║   ██║
  ╚══════╝╚═╝  ╚═══╝╚═╝  ╚═╝╚═╝      ╚═════╝ ╚══════╝╚═╝  ╚═══╝   ╚═╝

  > browser extension + local bridge for Roblox Studio, VS Code, and MCP
  > no API key · no terminal · no copy-paste

snapgentV3 is a browser extension plus a local bridge that connects a normal AI chat to Roblox Studio and VS Code. Use Roblox Studio's built-in MCP integration to read, edit, run, and inspect a Roblox project, or select the VS Code MCP server to work with a code workspace. Describe what you want in plain language — the extension forwards commands to the selected server through the local bridge, and returns the results to chat.

Free and open source — no API key, subscription, or sign-in required. DeepSeek is the recommended provider. Roblox Studio is the built-in MCP target; VS Code is an optional configured target. Available actions depend on the selected target.

01   How it works

┌─────────────┐     plain-text      ┌──────────────────┐    WebSocket    ┌────────────┐    MCP    ┌────────────────┐
│  AI chat    │ ───  commands  ───▶ │ snapgentV3       │ ──────────────▶ │  Bridge    │ ────────▶ │ Selected MCP   │
│  (browser)  │ ◀──  results   ───  │ browser extension│ ◀────────────── │ (local)    │ ◀──────── │ target         │
└─────────────┘                     └──────────────────┘                 └────────────┘           └────────────────┘

1. You describe a task in a supported AI chat.
2. The AI replies with a snapgentV3 command.
3. The extension detects the command and forwards it to the local bridge.
4. The bridge routes it to the selected Roblox Studio or VS Code MCP server.
5. The result is sent back into chat, and the AI continues working.

Choose the target in the snapgentV3 bar before starting a session. Roblox Studio is the built-in integration; configure the VS Code MCP server in the local bridge before selecting it.

02   Supported AI providers

Provider activation is configured in manifest.json (content_scripts and host_permissions) and PROVIDER_URLS in background.js.

03   Requirements

• Microsoft Edge or Google Chrome (Manifest V3).
• Roblox Studio with its built-in MCP server enabled when using the Roblox target.
• A VS Code MCP server configured in the local bridge when using the VS Code target.
• Windows or macOS for the bridge.
• Python 3.9+ on macOS. The Windows release includes the bridge executable.
• The bridge running while you use the agent.

04   Installation

① Load the extension

1. Download and extract the latest release.
2. Open edge://extensions (Edge) or chrome://extensions (Chrome).
3. Enable Developer mode.
4. Click Load unpacked and select the snapgent-extension folder.
5. The snapgentV3 icon appears in your browser toolbar.

② Set up the MCP target

For Roblox Studio, open a place and enable Assistant AI → … → Manage MCP Servers → Enable Studio as MCP Server.

For VS Code, add its MCP server configuration to your local config.json. Keep machine-specific executable and workspace paths in this local file; do not publish them.

③ Start the bridge

• Windows — double-click start.bat. It detects Roblox Studio MCP and
  creates or updates the local config.json automatically. Existing optional
  Existing local bridge settings are preserved.
• macOS — double-click MacOS_Start.command.

A window opens and stays open while the bridge runs. On macOS, the first launch may show a Gatekeeper warning; select Done, then allow the script under System Settings → Privacy & Security.

④ Start a session

1. Open a supported AI chat site and start a new chat.
2. Select Roblox Studio or VS Code from the snapgentV3 MCP target menu.
3. Click Start and describe what you want to make.

05   Usage

Once a session is running, just talk to the AI normally:

"Inspect the scripts in my place and tell me how the round system works."

"Build a shop UI with buttons for the items in ReplicatedStorage."

The AI sends commands, snapgentV3 runs them in Studio, and the results appear in chat. You can give the AI another instruction at any time.

Tips

• Keep the bridge window open — closing it disconnects the agent.
• Choose VS Code from the bar when you want to work with a code workspace instead of Roblox Studio.
• On Arena, use Direct mode.
• If Gemini or Kimi stops using tools in a long session, remind it to use the Roblox commands or start a new chat.

06   Architecture

The extension is split into a provider-agnostic core and per-site providers. The core talks to each AI site through the ZSProvider interface.

core/config.js              system prompt, feedback, tool categories
core/parser.js              snapgentV3 command parsing
core/main.js                agent loop, UI, session state
providers/<site>.js         site-specific selectors and chat integration
background.js               WebSocket connection to the local bridge
manifest.json               browser permissions and content scripts

The bridge routes commands to the Roblox Studio or VS Code MCP server selected in the chat bar.

07   Project structure

snapgent-full/
├── snapgent-extension/        # Load this folder as an unpacked extension
│   ├── core/                  # Shared prompt, parser, agent loop, and UI
│   ├── providers/             # Per-AI-site integrations
│   ├── background.js          # Local bridge connection
│   ├── manifest.json          # Manifest V3 configuration
│   └── test-*.js              # Smoke tests
├── assets/                    # Project images
├── setup-config.ps1           # Detects Roblox Studio MCP on Windows
├── config.json                # Generated locally by start.bat (not tracked)
├── start.bat                  # Windows bridge launcher
├── MacOS_Start.command        # macOS bridge launcher
├── bridge.exe                 # Windows local bridge
└── vscode_mcp.exe             # Optional VS Code MCP server

08   Development

Plain Node, no extra dependencies. Run from snapgent-extension/:

node test-parser.js
node test-chatgpt.js
node test-meta.js

Each test prints PASS/FAIL and exits with a non-zero status if a case fails.

Notes

• Keep site-specific DOM logic in providers/; the core should use ZSProvider.
• The bridge listens on local port 17613.
• Provider URLs must stay in sync across manifest.json and PROVIDER_URLS in background.js.

09   Troubleshooting