Sensei & agents · Lesson 6 of 6
How to use PixelDojo in Claude, ChatGPT and Cursor
Connect once with the PixelDojo connector, then ask your assistant for images, video, ads and whole films.
You need: Claude, ChatGPT, Cursor or another AI app that takes MCP.
Step by step
Add a custom connector
In Claude, open Settings, then Connectors, click Add and choose Add custom connector.
Paste pixeldojo.ai/mcp
Name it PixelDojo and paste https://pixeldojo.ai/mcp, then click Continue.
Sign in and approve
Sign in to PixelDojo when asked. No API keys, no code. You choose which tools need your approval.
Just ask
"Make me a cinematic portrait of a lighthouse keeper, then animate it." Your assistant picks the models and asks before it uses a tool.
Why it works
- Ask for outcomes, not models: "a vertical video ad for this product" gets a script, a presenter and a render.
- The same connector works in Claude, ChatGPT, Cursor, Codex and more. Set up instructions for each are on the Skills page.
Your practice task
Connect PixelDojo to your assistant and ask it for one image and one video in the same message.
Read the transcript(2:00)
Everything you've seen in these tutorials, every shot, every voice, even me, was made by asking an AI. That's the PixelDojo Connector, and it's easier to set up than a coffee machine. It works in Claude, ChatGPT, Cursor, Codex, and more. Here's Claude. Open Settings, then Connectors, click Add, and choose Add Custom Connector. Name it PixelDojo, paste pixeldojo.ai slash mcp, and click Continue. Sign in to PixelDojo when asked, and approve. No API keys, no code. And that's it. Claude now has the whole PixelDojo toolkit. Images and video, consistent characters, editing, upscaling, audio, ads, even whole films. You choose which ones need your approval. Now just ask. Make me a cinematic portrait of a lighthouse keeper, then animate it. Claude picks the right models and asks your permission before it uses a tool. A moment later, you have the portrait, and the moving shot. Give it a product photo, and ask for a video ad. It writes the script, casts a presenter, and renders a vertical ad, ready to post. Or describe a whole story. One sentence becomes a script, a storyboard, every shot, a soundtrack, and a finished film. It can even make a presenter who talks, face, voice, and lip sync from one message. Hi, I'm your new AI presenter. Everything you're seeing, my face, my voice, this studio, was made in one chat with PixelDojo. Rather how I was made, actually. Connect it once, then just ask. You'll find the setup for every app at pixeldojo.ai.