Build a sequence of shots from a shot list

Put the reference images in one collection. Then, for every row of the shot list, make the first frame (and the last frame where the row asks for one) with GPT Image 2.5 Sunburst (model_openai-gpt-image-2-5-sunburst), 16:9, with the reference images as referenceImages. Assemble the stills into one board with Grid Maker (model_scenario-grid-maker) and show it to me. Quote the cost with a dry run first and wait for my go. No clips until I approve the board.

For each approved row, in order: run the row's model (Minimax H3 is model_minimax-h3, Wan 3.0 is model_alibaba-wan-3-0) with the approved still as the first frame, and the last frame where the row has one (read model_schema_get for each model's parameter names). Prompt: start state, one action, end state, camera, sound. Duration from the row, a low resolution first, the aspect ratio following the first frame. Quote each run with a dry run first and wait for my go. To chain two shots, extract the last frame of the accepted clip with Video to Image Sequence (model_scenario-video-to-image-seq), show it to me at full size, and use it as the next first frame.

For the lounge shot (Wan 3.0, 5 seconds, 720P, the sunset still as the first frame and the night still as the last frame), the prompt word for word:
Time passes in the airport lounge. The low sunset light through the glass fades steadily over the whole shot into a dark blue night, the sky outside darkens, the runway and apron lights come on one by one, and the pendant lamps, the table lamps and the candles in the lounge glow warmer as the room darkens. The two travelers at the far table sit still, talking quietly. Nothing else in the room moves. Static camera, locked-off tripod shot, same framing from start to end. Sound: soft lounge room tone, the distant hum of a jet engine outside. No music, no speech.

Wait for every job with jobs_wait and show me the clips in order. Tell me which shots fail the checks and redo only those, quoting first.

Join the accepted clips in order with Scenario Video Concat (model_scenario-video-concat), after conforming them to one size and frame rate; read model_schema_get for its inputs and quote it first. Then add the stills, the accepted clips and the cut to the collection, and report the models, the settings and every asset ID by shot number.

The same airport lounge from the reference image, same camera position, same framing, same furniture, same people in the same places, earlier the same day at sunset: warm low sunlight streams through the floor-to-ceiling glass and lays long golden stripes across the floor and the cream leather armchairs, the sky outside is orange and pale blue, the parked aircraft is lit by the sun, the pendant lamps and the bar lights are switched off, the room is lit only by daylight. Keep every object, its position and its color exactly the same. Wide 16:9 frame. Keep the small NORTHVANE wordmark on the wall exactly as it is; no other text. Style: premium travel advertising, warm and calm, realistic materials.

Time passes in the airport lounge. The low sunset light through the glass fades steadily over the whole shot into a dark blue night, the sky outside darkens, the runway and apron lights come on one by one, and the pendant lamps, the table lamps and the candles in the lounge glow warmer as the room darkens. The two travelers at the far table sit still, talking quietly. Nothing else in the room moves. Static camera, locked-off tripod shot, same framing from start to end. Sound: soft lounge room tone, the distant hum of a jet engine outside. No music, no speech.