Hosted remote MCP server · OAuth 2.1

Drive AI image & video from any MCP client

ZenCreator's generation tools — text-to-image, reference editing, face-swap, upscale, lip-sync, and video — exposed as a hosted MCP server. Connect Claude, Cursor, ChatGPT, or Claude Code. No API keys to paste, nothing to install. Auth and credits use your existing ZenCreator account.

Server URL

https://mcp.zencreator.pro/mcp

Add the URL, log in once in your browser, and you're connected. Works with Claude · Cursor · ChatGPT · Claude Code.

Beta. The ZenCreator MCP server is in active beta. We'd be grateful for your comments and bug reports — they directly shape what ships next.

zencreator-mcp — tldr

# TL;DR

> ZenCreator MCP is a hosted, OAuth-secured Model Context Protocol server that lets Claude, Cursor, ChatGPT or Claude Code generate images and video through ZenCreator.

# what you get

  • Connect one server URL — no API keys to paste, nothing to install.
  • Text-to-image, reference editing, face-swap, image-to-video, lip-sync and upscaling.
  • Runs on your existing ZenCreator account and credits.

Connect your client

Pick your MCP client. Each adds the same server URL, then completes a one-time OAuth login.

Claude Code

  1. 1

    Add the MCP server globally:

    claude mcp add -t http -s user ZenCreator https://mcp.zencreator.pro/mcp
  2. 2

    Start Claude Code, then authenticate with ZenCreator:

    /mcp

Follow the authentication flow in your browser.

Full Claude Code guide
Video guide

How do you set up ZenCreator MCP?

Watch the full walkthrough — add one server URL, log in once, and run your first image or video generation from your agent in about two minutes.

What this unlocks

Workflows the dashboard can't do in one click.

Pipeline

Stitch tools in one prompt

Image → animation → lipsync → upscale. Intermediate assets stay in context — no save and re-upload between tools.

One clip. The dashboard is four tools, four uploads, four waits.

ZenCreatorvideo
Generate a 16:9 clip of a smiling barista, animate her waving, then lipsync her saying "Hi, welcome to Ferro — what'll it be today?"

Generated reel · 16:9

At scale

Personalize hundreds of assets

Iterate over a CSV, a Notion database, or a folder of photos with a single instruction.

Fifty hero shots in one command. The dashboard makes you submit one at a time.

ZenCreatorimage
For each product in products.csv, generate a hero shot in our brand style — soft lighting, white background, 16:9.
AI-generated 16:9 product hero shot on a soft white studio background, produced with ZenCreator MCP

Batch output · 50 hero shots · 16:9

Compose

Mix with the rest of your stack

Pull from Linear, Notion, Stripe, files — feed straight into generation, deliver back where your team works.

Generation triggered by real work, delivered where the team lives. Not possible from a browser tab.

ZenCreatorimage
Read this week's Linear ship list, generate a patch-notes header for each, and post the bundle to #releases.
AI-generated release header graphic in a violet gradient, produced with ZenCreator MCP

Release header · 16:9 · posted to #releases

People also ask

Hover or tap a card to reveal the answer.

What is ZenCreator MCP?

See the answer

ZenCreator MCP is a hosted Model Context Protocol server. It exposes ZenCreator’s image and video generation tools so any MCP-compatible agent can use them on your behalf.

What is the Model Context Protocol (MCP)?

See the answer

MCP is an open standard that lets AI agents connect to external tools and data. ZenCreator runs an MCP server so agents like Claude, Cursor and ChatGPT can generate through it.

How does ZenCreator MCP work?

See the answer

You add the ZenCreator server URL to your MCP client and sign in once with OAuth. Your agent then calls ZenCreator’s tools and gets finished images and video back in the chat.

Is ZenCreator MCP secure?

See the answer

Yes. The server uses OAuth 2.1 — you authenticate in your browser with your existing ZenCreator account, so no API keys are stored in the client.

Can one agent run a full generation pipeline?

See the answer

Yes. In a single prompt an agent can chain steps — generate an image, animate it, lip-sync, then upscale — with intermediate assets kept in context.

Frequently asked questions

Bring ZenCreator into your agent

Connect once — no API keys, on your existing plan credits — and generate images and video without leaving the chat.