Create images and video
from a ChatGPT conversation.
Connect the JsonCut MCP server so ChatGPT can create projects, use uploaded files, inspect frames, fill templates and render results.
JsonCut MCP gives compatible ChatGPT clients a focused V2 tool surface for iterative image and video authoring.
Connect JsonCut.
Render and deliver.
Connect JsonCut.
Render and deliver.
Connect JsonCut
Authenticate the integration and choose a project or versioned template.
Map data and media
Inspect the input contract, upload files and map only the values the design expects.
Render and deliver
Start a job, observe its status and pass the finished output to the next step.
A focused connection, not another disconnected tool.
Use chatgpt + mcp where it helps and keep the visual source understandable in Studio.
Keep long work structured
The agent can build scene by scene and patch only the relevant part of a long video.
Review what it created
Image snapshots and multi-frame video grids let the agent inspect the visible result.
Use templates when faster
Search existing templates, retrieve fields and fill them without rebuilding familiar work.
Conversational campaign creation.
And the work after it.
Before you connect it.
01Do I need to build the visual in code?
No. Create and approve it in Studio, then expose typed fields for the integration.
02Can the workflow generate images and video?
Yes. Both project types use the same media, template and job foundations.
03How are large media files supplied?
Upload them as project media first and use the returned reference instead of embedding large base64 strings.
Use the tools you know.
Keep control of the result.
Connect JsonCut where it belongs and keep the visual source accessible in Studio.
Start creating free