Asset library — generate once, reuse across the video
A video-level library of the characters, locations, products and the logo this reel needs — each defined once and assigned across the clips and shots that use it. Identity stays consistent from the first frame to the last, and nothing is re-made per clip. Demo brand: Fidelis Global · 30s client-proof reel.
Library 4 assets · assigned across 4 clips
Ready
Presenter
Ops presenter
Ready
Office
Office — open plan
Dashboard
Fidelis dashboard
Ready
Logo
Brand logo
New asset
Source an asset Selected · Fidelis dashboard
Fidelis dashboard
Product
The product UI featured in the Turn and Payoff clips. Define it once here and every shot that needs the dashboard references this one asset — so it looks identical in each.
Already have it / upload
Drop in a real screenshot or export of the dashboard — the truest source when the product already exists.
Generate in-app
Make it with the same chat-iterative image engine as image posts — refine until it’s right, then it’s reusable.
Get a prompt
Copy a ready-to-paste prompt to generate the asset in another tool, then upload the result back here.
Composite
Combine library assets — e.g. place a character into a scene — to produce the exact frame a shot needs.
Generate in-app · Fidelis dashboard
Dashboard
Dashboard
Dashboard
Same image engine as image posts. Generate once → the chosen image becomes the reusable “Fidelis dashboard” asset, referenced by every clip that needs it.
Composite — place a character in a scene
Character · Ops presenter
Presenter
+
Scene · Office
Office
→
Combined frame
Produces a usable first/last frame with the character placed in the scene — the actual image a clip needs, built from two library assets so identity and setting both stay on-model.
Video-level reuse. A character or location made once stays consistent across every clip and shot that references it — no identity drift between frames, and no credits wasted re-generating the same thing per clip.
Three ways to source each asset. Already have it (upload the real thing), generate it in-app with the image engine, or get a prompt to make it in another tool — whichever fits the asset.
Composite makes the real frame. It places a library character into a library scene to produce the exact first/last frame a clip needs — assets in, a usable keyframe out.