A prompt-writing method for image and video: storyboard briefs over adjective stacking

On November 4, 2026, the AI section of DEV Community published "Writing Image and Video Prompts as a Small Production Brief", laying out a prompting approach that applies to both image and video generation. The author's central suggestion is that a prompt should read like a mini production brief rather than a pile of adjectives. Whether the creator is working on a portrait, a fashion edit, a festival greeting, or a short video concept, the same sequence helps: define the subject, describe the scene, set the visual mood, and then add one controlled movement or camera cue. The first step is to lock down whatever must stay consistent. That means naming the subject, the environment, the time of day, and the intended point of visual focus in the frame. Only after that does the author move on to the finer layers one by one: texture, color, the direction of light, the feel of the lens, and the composition. The final layer is a single controlled movement or camera instruction appended at the end. On iteration, the advice is conservative: when a first result is already close to what was wanted, change only one variable and then revise again. As of now this topic has only this one report, with no follow-up developments published yet.