The idea is:
The ‘Retrieve an execution’ api call with includeData=true can return masses of input and output data, much of which might be irrelevant to the purpose of the call. It would be flippin’ great to be able to specify one or nodes for which the input and output data is relevant so that the response payload can be kept as small as possible without compromising its utility.
My use case:
I’ve got n8n hooked up to some AI dev tools to support me with workflow development and debugging. The size of the execution response payloads is making my prompt context enormous, which is costly.
I think it would be beneficial to add this because:
- It would make AI development assistance more cost-effective
- It would anyway help to limit the size of response payloads
- There should always be a third bullet point
Any resources to support this?
Are you willing to work on this?
It’s beyond my ability to help develop this functionality but I’d be very happy to test and support.