The Kind Princess — Story Video · skill pubblica Open Skill
The Kind Princess — Story Video
The Kind Princess — Story Video è una Open Skill pubblica su Digen. Aprila per vedere cosa crea questo workflow.
Anime
1 1 5
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/3761876/2026-09-26/utfb7r55.png
# The Kind Princess — Story Video ## When to use - The user wants to produce the full 2-minute "The Kind Princess" story video: 12 scenes × exactly 10 seconds, 16:9, whimsical hand-painted anime fantasy, family-friendly, moral about kindness. - The user wants to advance or change one step: regenerate/restyle a scene, redo a character, produce the final mixed video, add voiceover/music, or generate the cover page. ## Reference files (read on demand, don't read all at once) - `references/characters.md` — Master character lock (Elara, King Arthur, Queen Sophia, Prince Leo, Mia, Snowy). Read before generating any image or video. - `references/scenes.md` — The 12 scene-by-scene prompts plus the cover page prompt. Read when generating scene images/videos or the cover. - `references/style.md` — Visual style, continuity, camera, subtitle, audio, negative prompt, and final quality check. Read when writing prompts. ## Steps 1. Confirm scope: the full video, or a single stage. For multi-step work, first call `write_todos` to list the steps and update the whole list as you progress. 2. Read `references/characters.md` and `references/style.md`. 3. Generate a character reference image for Elara with `generate_image` (`model='t2i'`, `orientation='landscape'`, `resolution='1080P'`, `name='Elara_ref'`, `entity_type='characters'`). Use this reference as `input_urls` for scene images where she appears to keep her consistent. 4. Generate the 12 scene storyboard images (`model='i2i'` with the Elara reference as `input_urls`, or `model='t2i'` if no reference; `orientation='landscape'`, `resolution='1080P'`, `name='scene_01'`…`'scene_12'`, `entity_type='storyboard_images'`) using the prompts in `references/scenes.md`. 5. Generate each scene video with `generate_video` (`model='i2v'`, `input_urls=[scene image]`, `orientation='landscape'`, `resolution='1080P'`, `duration=10`, `name='clip_01'`…`'clip_12'`, `entity_type='storyboard_videos'`). 6. Generate voiceover for the dialogue lines with `generate_tts` (`model='tts.design'`, `voice_instruction` describing Elara's clear, gentle, youthful female voice; `name='vo_<scene>'`, `entity_type='voiceovers'`). 7. Generate background music with `generate_music` (`model='music'`, prompt describing warm magical orchestral/piano that builds during adventure and resolves upliftingly; `name='soundtrack'`, `entity_type='music'`). 8. Mix the 12 clips plus voiceover and music into the final video with `mix_clips` (`clips=[...]`, `orientation='landscape'`, `name='the_kind_princess_final'`, `entity_type='final_videos'`). 9. Generate the cover page (`model='t2i'`, `orientation='portrait'`, `name='cover'`, `entity_type='storyboard_images'`) using the cover prompt in `references/scenes.md`. 10. Summarize results in natural language; call `set_guidance` to suggest next steps. ## Notes - Keep every character 100% consistent across all scenes (face, hair, eyes, dress, tiara, voice); never change age, hairstyle, clothing, or body proportions. - Every scene is exactly 10 seconds; total duration ≈ 2 minutes. - Use the whimsical hand-painted anime fantasy aesthetic; avoid photorealism, generic 3D toy rendering, and copying any specific copyrighted character or film. - Video generation does not support NSFW/adult content on any channel. - Report failures honestly and never fabricate URLs. - Intermediate storyboard images/videos will be visible on the canvas — that is expected.