The idea is:
Add a “Pin output data” shortcut button to the node hover action bar (the toolbar that appears when hovering over a node, currently showing Execute, Disable, and Delete). Clicking it would immediately pin the most recent execution output of that node — without requiring the user to open the node editor.
If no output data is available yet (node hasn’t been run), the button could be grayed out or hidden.
My use case:
When building and testing long pipelines that include multiple AI nodes (e.g., LLM calls, embeddings, summarization chains), I frequently need to iterate on the later stages of the pipeline without re-running the expensive upstream nodes every time.
Currently, to pin a node’s output I have to:
Double-click to open the node
Scroll to find the pin icon in the output panel
Click it
Close the node
With three or four AI nodes in a pipeline, this becomes repetitive and slow — and easy to forget, causing unnecessary API calls and token usage on every test run.
I think it would be beneficial to add this because:
It significantly speeds up iterative testing with real data on complex pipelines
It reduces unnecessary re-execution of costly nodes (AI/LLM calls, external API requests, slow database queries) during downstream development
The hover shortcut bar is already the established UX pattern for quick node actions — pinning output is a natural and frequent workflow action that fits right in
It would make working with real data in long pipelines much less friction-heavy, without changing any existing behavior
Any resources to support this?
The existing hover shortcut bar UX pattern is already present in the canvas — this extends it consistently
Are you willing to work on this?
No, I’m not a contributor to the n8n codebase — but I’d be happy to provide detailed UX specs or help test any implementation. Don’t forget to upvote if this would help your workflow too!