Connect SVGator to your AI assistant

The SVGator connector plugs SVGator into an AI assistant, so you can browse, edit, and export your projects by chatting instead of clicking through the editor. Because it follows the open MCP standard, it works with any AI assistant that can add a custom MCP server.

Table of contents

Connect from Claude

Connect from ChatGPT

Connect from Gemini

Connect from Grok

Connect from Cursor

Connect from VS Code

Other tools

Connect from Claude

Add from the connector directory 

In claude.ai or Claude Desktop, open Settings and go to Connectors, then click Browse connectors.

Find SVGator in the directory (you can search for it by name) and click Connect.

Add as a custom connector

Custom connectors are available on Claude Pro, Max, Team, and Enterprise plans.

In claude.ai or Claude Desktop, open Settings and go to Connectors, then choose Add custom connector. 

Enter SVGator as the name and https://mcp.svgator.com/mcp as the URL, then click Add.

Claude opens a browser window where you sign in to SVGator and approve access. Once you approve, the connector shows as Connected, and the SVGator tools are ready to use in your chats.

Working in Claude Code instead? Add the connector from the command line, then run /mcp and pick svgator to sign in:

claude mcp add --transport http svgator https://mcp.svgator.com/mcp

What you're approving and what Claude can see

When you approve access, you're letting Claude work with your SVGator projects on your behalf: listing them, opening them, editing them, and exporting them.

Your sign-in runs through SVGator's normal login, and your credentials, like your password and access token, stay on SVGator's servers. Claude never sees them.

What Claude does see is limited to the results of what you ask for: your list of projects, summaries of a project's structure, the values you have it read, and any files or links it exports. Nothing else about your account leaves SVGator.

When the connection expires

If the connection ever gets stuck or stops responding, remove the SVGator connector under Settings → Connectors in Claude, add it again with the same details, and approve access once more. Removing the connector only revokes Claude's access; your SVGator account and projects are never affected.

Connect from ChatGPT

Add the SVGator connector to ChatGPT to browse, edit, and export your projects by chatting.

In ChatGPT, open Plugins from the sidebar, click Add, then select Add custom MCP server.

Enter SVGator as the name, https://mcp.svgator.com/mcp as the server URL, and OAuth as the authentication, then click Create as a plugin.

Click Connect SVGator, then sign in to SVGator and approve access.

In a new chat, select SVGator from the tools menu.

Connect from Gemini

Add the SVGator connector to Gemini to browse, edit, and export your projects by chatting.

At gemini.google.com, open Customize, go to Connectors, then click Custom.

Under Connect to an MCP server, enter https://mcp.svgator.com/mcp as the MCP server URL, then click Next.

Review the security and privacy notice, then click Connect.

In the window that links your Google account and SVGator account, click Agree and continue.

Connect from Grok

Set up the SVGator connector in Grok, then ask it to list, edit, or export your animation projects without opening the editor.

At grok.com, go to Plugins, then click New Connector in the top-right corner.

Select Custom.

Enter SVGator as the name and https://mcp.svgator.com/mcp as the server URL, then click Add Connector.

Connect from Cursor

Add the SVGator connector to Cursor's MCP servers and manage your animation projects from the same place you write code.

In Cursor, click Customize.

Open the MCPs tab.

Click New MCP Server.

Add the SVGator configuration below to the JSON file, then save it with Cmd + S (Ctrl + S on Windows).

{  "mcpServers": {    "svgator": {      "url": "https://mcp.svgator.com/mcp"    }  }}

Once the file is saved, SVGator appears in the MCP list. Click Authenticate.

Sign in to SVGator and authorize Cursor's access to your account.

Connect from VS Code

Open the Command Palette (Ctrl + Shift + P, or Cmd + Shift + P on Mac) and run MCP: Add Server.

Select HTTP, then enter https://mcp.svgator.com/mcp as the server URL.

The first time you use the SVGator MCP, VS Code asks you to authenticate. You can also authenticate through Copilot.

Other tools

Using an AI client that isn't covered in the other guides? The SVGator connector works with it too, as long as it meets the requirements below.

Client requirements

Use https://mcp.svgator.com/mcp as the endpoint in any client that supports remote MCP servers over Streamable HTTP and OAuth 2.1 with Dynamic Client Registration and PKCE. The client must support MCP tools; support for resources and prompts is optional.

Configuration formats

Clients commonly use one of these two configuration formats:

{  "mcpServers": {    "svgator": { "url": "https://mcp.svgator.com/mcp" }  }}

{  "servers": {    "svgator": {      "type": "http",      "url": "https://mcp.svgator.com/mcp"    }  }}

Limits

Clients that support only stdio, legacy SSE, static API-key headers, or no browser OAuth can't connect directly. Client plans and organization policies may also block remote or write-capable MCP servers.