Visualizee MCP server

Renders and video,right inside your AI assistant.

Connect Visualizee to Claude, ChatGPT, Lovable, Claude Code or Cursor once. Then attach a photo, describe the shot and get a client-ready image or video back in the chat.

Try 5 free renders Connect your client

Render my SketchUp model as a real house at golden hour. Keep the design and the camera.

Used Visualizee

Pro quality

Here it is at golden hour, same design and camera. Want a blue hour version with the lights on?

Connect the agent you already use.

Visualizee is a standard remote MCP server. Pick your agent, add the server URL and sign in once with your Visualizee account. No API keys.

Claude

Add the connector

In claude.ai or Claude Desktop, go to Settings → Connectors and choose Add custom connector. Name it Visualizee and paste the server URL:

https://visualizee.app/api/mcp

Sign in once

Click Connect. Your browser opens visualizee.app: sign in and approve access. Then enable the connector in a chat and ask for a render.

Claude Code

Add the server

Run this once in your terminal:

claude mcp add --transport http visualizee https://visualizee.app/api/mcp

Authenticate

Inside a session, run /mcp, pick visualizee, and sign in when the browser opens. The tools are then available in every session.

ChatGPT

Create the connector

In ChatGPT, go to Settings → Apps & Connectors, enable Developer mode under Advanced settings, then choose Create and paste the server URL:

https://visualizee.app/api/mcp

Sign in once

Pick OAuth as the authentication method and connect. You sign in at visualizee.app and approve access. Enable the connector in a chat and ask for a render.

Lovable

Add the connector

In Lovable, open the Connectors dashboard, scroll to the bottom of the All view, and pick the Custom (MCP) card. Name it Visualizee and paste the server URL:

https://visualizee.app/api/mcp

Authorize

Keep OAuth selected and click Add & authorize, then sign in at visualizee.app and approve access. Then ask for a render in the Lovable chat.

Cursor

Add the server

Add this to ~/.cursor/mcp.json (global) or .cursor/mcp.json in your project:

{
  "mcpServers": {
    "visualizee": {
      "url": "https://visualizee.app/api/mcp"
    }
  }
}

Authenticate

Cursor asks you to sign in on first use. Approve access at visualizee.app and you are done.

Codex

Add the server

Run this once in your terminal. In the Codex app or IDE extension, use Settings → MCP servers → Add server, pick Streamable HTTP and paste the server URL instead.

codex mcp add visualizee --url https://visualizee.app/api/mcp

Sign in once

Run codex mcp login visualizee (or click Authenticate in the app). Sign in at visualizee.app, approve access, and ask Codex for a render.

Gemini

Add the server

In Gemini CLI, run this once. In the Gemini web app, go to Settings → Connected apps → Add a custom app and paste the server URL instead.

gemini mcp add --transport http visualizee https://visualizee.app/api/mcp

Sign in once

Your browser opens visualizee.app on first use: sign in and approve access. If it does not, run /mcp auth visualizee inside Gemini CLI.

Grok

Add the connector

On grok.com, open Connectors → New connector → Custom and paste the server URL. In the Grok Build CLI, run this instead:

grok mcp add --transport http visualizee https://visualizee.app/api/mcp

Sign in once

Grok opens visualizee.app on first use: sign in and approve access. Then enable the connector in a chat and ask for a render.

Perplexity

Add the connector

On the web or desktop app, go to Settings → Connectors → Custom connector, choose Remote, name it Visualizee and paste the server URL:

https://visualizee.app/api/mcp

Sign in once

Pick OAuth as the authentication and click the connector card. Sign in at visualizee.app and approve access. Custom connectors need a Pro or Max plan.

Manus

Add the connector

In Manus, open Connectors → Create → Add MCP by URL. Name it Visualizee and paste the server URL:

https://visualizee.app/api/mcp

Sign in once

Choose OAuth when Manus asks how to authenticate, then sign in at visualizee.app and approve access. Ask Manus for a render.

OpenClaw

Add the server

Run this once, then set auth: "oauth" under mcp.servers.visualizee in your OpenClaw config:

openclaw mcp add visualizee --url https://visualizee.app/api/mcp --transport streamable-http

Sign in once

Run openclaw mcp login visualizee, open the link it prints, sign in at visualizee.app and approve access. Restart the gateway if it is already running.

Hermes Agent

Add the server

Add this to ~/.hermes/config.yaml, then run /reload-mcp:

mcp_servers:
  visualizee:
    url: "https://visualizee.app/api/mcp"
    auth: oauth

Sign in once

Hermes opens visualizee.app on first connect: sign in and approve access. On a server without a browser, run hermes mcp login visualizee --flow device.

Kimi

Add the server

In Kimi Code, add this to ~/.kimi-code/mcp.json (or run /mcp-config in a session) and restart:

{
  "mcpServers": {
    "visualizee": {
      "url": "https://visualizee.app/api/mcp"
    }
  }
}

Sign in once

Run /mcp-config login visualizee. Your browser opens visualizee.app: sign in, approve access and ask Kimi for a render.

Other MCP client

Add the server

Any client that supports remote MCP servers over Streamable HTTP with OAuth works. Add a server named visualizee with this URL:

https://visualizee.app/api/mcp

Sign in once

The client opens visualizee.app when it first connects: sign in and approve access. No API key or token to paste.

Read the developer docs

What your assistant can do.

The same renders, qualities and workspace as the app, from the chat you already use.

Renders in the chat
Ask for a shot in plain words. The result comes back as a preview in the chat, plus a link to the full image.
Your photos as input
Attach a photo of a kitchen, a car or a product, or a model screenshot. The assistant uploads it and Visualizee keeps it as the base of the render.
Edit by iterating
Change only the wheels, only the wall color, only the sky. Each render can feed the next one, so a series stays consistent.
Video, not just stills
Turn a render into a clip with a camera move. Your assistant confirms the length, the resolution and how many renders it will use before it starts.
Saved to your workspace
Every render lands in your Visualizee workspace, in files and folders you name from the chat. Same account and same plan as the app.
Sign in once
OAuth with your Visualizee account. No API keys to copy or keep safe.

Renders and video,right inside your AI assistant.

Connect the agent you already use.

Add the connector

Sign in once

Add the server

Authenticate

Create the connector

Sign in once

Add the connector

Authorize

Add the server

Authenticate

Add the server

Sign in once

Add the server

Sign in once

Add the connector

Sign in once

Add the connector

Sign in once

Add the connector

Sign in once

Add the server

Sign in once

Add the server

Sign in once

Add the server

Sign in once

Add the server

Sign in once

What your assistant can do.

Renders in the chat

Your photos as input

Edit by iterating

Video, not just stills

Saved to your workspace

Sign in once

Render from your assistant.