Connect data to creative.Not just to a rigid slot.
Build editable image and video systems, inspect their inputs and run them from n8n, Make, your application or an AI agent.

Map the useful inputs.Keep the output visible.
JsonCut gives automation builders a controlled template path and a flexible project path on the same rendering foundation.
Start with a reviewed template
Discover the exact typed fields before wiring a production workflow.
Upload once, reuse safely
Project media references replace repeated base64 payloads and unclear external links.
Handle real outcomes
Async jobs, webhooks and row-level batch states support reliable automation.
Made in JsonCut.Open one to see the project behind it.
Every title, clip and caption in these is still a layer. Open the project and change whatever you like.

Service Maintenance Alert
Open this example: Service Maintenance Alert
Onboarding Map
Open this example: Onboarding Map
West Pier maintenance board
Open this example: West Pier maintenance boardRead the fields.Route the result.
- Step 1
Read the fields
Load the template fields and media requirements before mapping a workflow.
- Step 2
Run in the background
Start one render or a batch from n8n, Make, the API or MCP.
- Step 3
Route the result
Use status and webhooks to store, publish, review or retry.
What you can make.
- Feed-to-creative workflows
- CRM personalization
- Agentic content operations
Before you start.The questions we hear most.
Can non-technical users work in JsonCut?
Yes. Studio is the primary visual workflow; API and MCP are additional paths, not prerequisites.
Can developers use the same projects?
Yes. Projects and templates created visually can be inspected and rendered through the API.
Are AI results still editable?
Yes. The assistant works with normal project layers and restorable versions rather than returning a flattened result.