What should a multi-reference fan edit disclose?

When a fan edit or photography experiment uses several reference images, what information makes the result easy to understand without turning the post into a technical log?

I am thinking about edits where one image supplies the main subject, another suggests the background or colour mood, and a third is only a style reference. A short note could record:

- which image was the main subject and what the other references contributed
- what the editor intended to change and what should remain recognisable
- whether a background, object, clothing detail, face, logo, or text was generated or materially changed
- whether the result is a creative composition rather than an untouched photograph
- that the source images were created by the poster or used with permission
- that the originals were kept so the edit can be compared with its sources

For fan and entertainment communities, the boundaries matter even more. I would avoid private member photos, unlicensed images, or morphing a real person's face or body. I would also avoid presenting a generated scene as evidence of something that happened, or using an edit to mock or make claims about an actor, character, or member.

The exact generation model or settings can change, and an output is not guaranteed to preserve identity, text, logos, small details, or the intended composition. Reviewing the full image and close details still seems necessary before sharing it.

What is the most useful minimum disclosure for a multi-reference fan edit: a simple edited/generated label, a list of reference roles, a before-and-after set, or a short change-and-preserve note?

Disclosure: I work on reference-image generation workflows, but I am intentionally not naming or linking a product here. This is a question about respectful editing, provenance, and clear presentation.