Browse documentation

Visualizee MCP server

Connect Visualizee to Claude, Claude Code, Cursor, or any MCP client. Render images by chat with your own account, credits, and workspace.

Visualizee ships a hosted MCP (Model Context Protocol) server, so any MCP-capable assistant, including Claude, Claude Code, and Cursor, can render images with your Visualizee account. Your assistant gets three tools: it can generate renders, look up the available models, and check your credits. Everything runs against your own account, and every render is filed into your Visualizee workspace.

The server lives at a single URL:

https://visualizee.app/api/mcp

It speaks Streamable HTTP and authorizes with OAuth 2.1: the first time a client connects, you sign in at visualizee.app and approve access. There are no API keys.

What this is good for

Blog and marketing imagery generated straight from your writing tool, renders dropped into a proposal while you draft it in Claude, or agent workflows that produce visuals on a schedule. If your assistant can call tools, it can now render.

Connect from Claude Code

Add the server from your terminal:

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

Then, inside a Claude Code session, run /mcp and pick visualizee to authenticate. Your browser opens visualizee.app; sign in and approve access. From that point the tools are available in every session.

If you prefer project-level config, add this to .mcp.json in your project root:

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

Connect from claude.ai and Claude Desktop

Open connector settings

Go to SettingsConnectors and choose Add custom connector.

Add the server

Name it Visualizee and paste https://visualizee.app/api/mcp as the URL.

Sign in

Click Connect. You will be sent to visualizee.app to sign in and approve access, then returned to Claude. Enable the connector in a chat and ask for a render.

Connect from Cursor

Add the server to .cursor/mcp.json (per project) or ~/.cursor/mcp.json (global):

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

Cursor will prompt you to authenticate on first use. Any other MCP client that supports Streamable HTTP and OAuth works the same way: point it at the URL and sign in when asked.

The tools

ToolWhat it does
render_image

Generates an image, waits for it to finish (up to about 4 minutes), and returns the final image URL plus a link to the file in your workspace.

list_models

Lists the available image models with their categories, credit costs, and supported aspect ratios and resolutions.

get_creditsReports the remaining credits, current plan, and total renders on the connected account.

You rarely call these by name. Ask for what you want ("render a 16:9 image of ...") and the assistant picks the right tool and parameters.

render_image parameters

ParameterDefaultNotes
promptrequiredPlain-language description of the image.
modelIdbest base modelAn id from list_models. Only set it when you want a specific model.
aspectRatio1:1For example 16:9, 4:3, 9:16. Must be supported by the model.
resolution1024Long-edge scale, for example 1024 or 2048.
ingredientUrlsnoneUp to 4 reference image URLs for image-to-image.
fileNameClaude MCP rendersThe Visualizee file the render is saved into, found or created by name.
folderNameworkspace rootA top-level folder for that file, found or created by name.

Where renders end up

Every MCP render is saved into your Visualizee workspace, not just returned as a URL. The fileName and folderName parameters control where:

  • With no names given, renders collect in a file called Claude MCP renders at the workspace root.
  • Pass a fileName per logical group, for example one file per blog post or per project, and a folderName such as "Blog posts" to keep things tidy.
  • Names match case-insensitively and are created on the fly, so repeated renders for the same job always land in the same file, even when several run in parallel.

Each result includes a direct link to the file, so you can jump from the chat into the workspace and keep iterating there with Vizzy.

Credits and limits

  • Renders cost the same credits as in the app; the cost and remaining balance are reported with every result. See plans & credits for what your tier includes.
  • If a render fails or cannot start, the credits are refunded automatically.
  • Pro-tier models respect your plan's monthly pro-render limit. When it is reached, the tool tells you and suggests a base model.
  • The server currently exposes image models only. Motion (video) stays in the app for now.
Long renders

render_image waits for the result and can take a few minutes at high resolutions. If a render times out but completes later, the image still appears in the target file in your workspace, and failed renders refund on their own.

Example prompts

Things you can ask once the server is connected:

  • "Render a 16:9 hero image of a Scandinavian living room at golden hour and save it in folder 'Blog posts', file 'Autumn palettes'."
  • "List the Visualizee models and pick the cheapest one that supports 21:9."
  • "Use this product photo as a reference and render it in a minimalist concrete showroom, 4:3."
  • "How many render credits do I have left?"

Next