AI video prompts: one shot per prompt

You fix three things first: where the viewer is, how many seconds you get, what they do after. Then the shots. The first second is not an opening, it is a filter.

Task: Text-to-video clips that move incoherently.

[Before you start, do three things]
1. Restate the goal in one sentence: who reads the output, and what decision it serves.
2. If I have not given you the following, ask me. Do not assume:
   - tool and duration limit
   - frame and action
   - continuous shot needed
   - aspect
3. Say how you plan to do it (at most 5 steps) and wait for my go-ahead.

[Go through each of these]
- one shot and one action per prompt
- subject, action, environment, light, camera, duration
- never combine complex action, multiple people and camera movement
- negatives: warping, extra limbs, flicker

[Process]
1. State platform, duration, viewing context and the target action
2. An opening that works in three seconds
3. Shot table: number, duration, frame, line or subtitle, sound
4. Mark the shot without which it fails
5. Two opening options

[Output]
Shot table with number, duration, frame, line, sound; the essential shot marked; alternate openings.

[Do not]
- No 'hi everyone, I am' openings
- Never exceed the shot limit
- Do not write unshootable shots
- No written-style dialogue