MiniMax H3 Prompt Writing · Open Skill 公開技能
MiniMax H3 Prompt Writing
MiniMax H3 Prompt Writing 是 Digen 上的公開 Open Skill。開啟後即可了解此工作流能製作什麼。
MV
533 30 123
https://d7qlwi7tl2u27.cloudfront.net/skill-covers/prod/2166889/2026-08-18/mbwgeeq9.png
# H3 Prompt Writing ## When to use - The user wants to generate a MiniMax H3 video (T2VA / I2VA / FL2VA / L2VA / Ref2VA) and needs a properly structured H3 prompt. - The user provides a multimodal request — text, reference images, videos, or audio — and wants it rewritten into the official H3 prompt format. - The user wants keyframes aligned, reference labels defined, or an existing H3 prompt checked/rewritten. ## Steps 1. Identify the input mode: T2VA, I2VA, FL2VA, L2VA, or full-reference Ref2VA. 2. For base text/keyframe modes, read `references/base-en.txt` and follow its final prompt structure. 3. For full-reference mode, read `references/ref-en.txt` and follow its six-section rewrite format. 4. Preserve the exact field names, section order, labels, and timing notation from the selected guide. ## Base Modes - T2VA: build the full audiovisual timeline from text. - I2VA: start from the first frame and develop forward from it. - FL2VA: describe the continuous path between the first and last frames. - L2VA: infer a plausible opening and converge to the supplied last frame. - Use `integrated_multimodal_description`, `overall_soundscape`, and `non_diegetic_music` in the order shown in `references/base-en.txt`. ## Full-Reference Mode - Ref2VA rewrites use `subject_definitions`, `summary`, `retention_analysis`, `detailed_description`, `overall_soundscape`, and `non_diegetic_music` in that order. Reference labels stay consistent across all sections. - Read `references/ref-en.txt` for label rules, retention analysis, and complete examples. ## Output Rules - Write rewrite sections in English; preserve dialogue, lyrics, and visible scene text in their original language. - Describe each shot by composition, subjects, environment, actions, camera, sound, and the exact point where referenced content appears. - Avoid plot summaries, unresolved reference labels, and timing that does not match the requested duration. ## Tips for Better Results - Always match the total duration of the description to the requested video length (4–15 seconds). - Keep reference labels consistent (e.g. <Subject 1>, <Picture 1>, <Video 1>, <Audio 1>) across every section. - Prefer concrete visual and audio details over abstract words like "cinematic" or "beautiful". - When using keyframes (I2VA / FL2VA / L2VA), clearly state how the first and/or last frame connects to the timeline. ## Notes - This skill only writes/rewrites prompt text; it does not call any generation or editing tools. - The reference files are an adaptation of MiniMax's official H3 prompt writing guides (github.com/MiniMax-AI/MiniMax-H3). Keep the field names, section order, labels, and timing notation exactly as written there. - If the user's request is ambiguous about the mode, ask which input they have (text only, first frame, first+last frames, last frame only, or full references) before rewriting. - Do not invent reference labels that were not provided; if a label is unresolved, tell the user rather than guessing.