Writes structured Seedance video-generation prompts with image-reference anchoring, timecoded shot lists, identity-continuity rules, cinematic color grading, and natural-sound direction. Use when the user asks for help with Seedance prompts, AI video prompts, text-to-video scene descriptions, cinematic video prompts, or any request to script a short generated film shot-by-shot. Trigger whenever the user mentions Seedance, video generation, text-to-video, AI video, shot list, scene prompt, or wants to turn a still concept into a timed video brief — even if they don't say "Seedance" by name.
91
88%
Does it follow best practices?
Impact
96%
2.28xAverage score across 3 eval scenarios
Passed
No known issues
Seedance is a reference-driven text-to-video model: it takes a written prompt plus one or two reference images and produces a short clip. This skill turns a rough scene idea into a production-grade Seedance prompt — image-anchored, timecoded, identity-stable, and cinematically directed.
Every prompt this skill produces has six layers. Include all six.
<<<image_n>>> and state explicitly what it locks (appearance, wardrobe, color grade, environment, face, lighting).[m:ss] blocks describing camera, subject action, framing, and on-screen sound at each beat. Total length 10–15 s.<<<image_n>>> declared at the top is doing real work in the shot list.A complete prompt that illustrates all six layers. Study the structure; don't ship it verbatim.
<<<image_1>>> is the wide shot reference — match the archer's exact
appearance, white kit, Olympic recurve bow, teal-to-black color grade,
dark reflective floor, and rim lighting.
<<<image_2>>> is the close-up reference — match the archer's exact face,
ocean blue eye, draw hand fingers, and bowstring detail. Same person
throughout.
[0:00]
Close-up of the archer's face in strict side profile. Ocean blue eye
locked forward on the target, unblinking. The taut bowstring runs
vertically beside her eye. Cold, focused, empty expression. Bowstring
creaking under maximum tension.
[0:05]
Camera drifts slowly down to her draw hand fingers gripping the bowstring
at the corner of her mouth — three fingers, knuckles white with tension,
every tendon visible. The arrow nock sharp in frame. The string at
absolute maximum load.
[0:09]
The release. Fingers open cleanly and instantly. Bowstring snaps forward
with explosive force. Audio: sharp crack of the release breaking the
silence.
[0:10]
Camera follows the arrow in flight across the dark void toward the target
— the arrow a thin dark line cutting through the teal atmosphere,
fletching spinning, shaft flexing.
[0:13]
Arrow strikes the exact gold centre of the target. Impact. Shaft quivers
violently. Audio: sharp deep thud of impact.
[0:15]
Camera holds on the arrow buried perfectly in gold, shaft vibration
slowly settling to stillness. Silence.
Stable identity throughout — same face, same hair, same white kit, same
bow in every moment. No distortion, no flickering, no ghosting. No music,
no crowd. Natural sound only — bowstring creak, crack of release, arrow
hiss, impact thud, silence. Teal-to-black cinematic color grade throughout.
4K ultra-high definition. Cinematic quality. This is a film not an
advertisement.78df912
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.