--- name: musicvideo-shots description: Write and generate every part of the music video — H3 Max Reference to Video with the face board, the look sheet, four frames and the part of the track as Audio 1, a JSON prompt of eight shots with cuts on beats, a different transition each, light, acting, head movement and lip-sync. --- # The prompt for every part of the video FlashBoards user skill, version 1.2.3 (2026-10-13). On the board, read these guides with `list_skills` and `get_skill` when the assistant exposes them; otherwise use the hosted index through an available read-only retrieval tool. Use that runtime’s actual model discovery, current page context and guarded board actions for execution. Keep native readiness, approval and cost gates; do not claim a guide was read when retrieval is unavailable. Read related guides with `get_skill({skillId:""})` or the [hosted index](https://pub-78da82d51ed14147ac554d38094b0c0b.r2.dev/user-skills/1.2.3/index.html). Reading a guide does not start or authorize a paid generation. Read `musicvideo` for the shared rules. References are the person's own {face board}, {look sheet} and the frames replaced for this story. > **Quality route:** Seedance 2.5 (`seedance-2-5`) with the same pictures as @Image1… and the part as @Audio1 does this job much better; quote both. ## Request `generate_video` with `minimax-h3-max-reference-to-video` · `duration` = the part rounded up · `resolution: "768P"` · `aspectRatio: "16:9"` · `referenceGenerationIds`: [{face board}, {look sheet}, frame 1…4] · `referenceAudioUrls`: [the part of {track}, its `audioFinishingTaskId`]. Start all parts together, then check each once. ## Prompt structure (JSON) - `output`: "A cinematic 16:9 music video told as a short film in three acts, part N of M ({act} and what happens). 8 shots with cuts and transitions on the beat at the given times, D seconds, then the video ends." - `style`: cinematic and natural, moving camera, the real light of each place, natural skin and sweat, real acting; no text, no screens, **no phones**, no slow motion. - `references`: Image 1 "only {character}'s face and hair, seen from five sides so they can turn their head freely; never on screen"; Image 2 "only their outfit [and prop]"; Image k "the place of shots 2j+1 and 2j+2: place, light, atmosphere and the people around; never its framing"; Audio 1 "the soundtrack… only {character} [raps/sings] the words". - `character`: "never frozen: their head, eyes and face move all the time and their expression changes with every moment". - `rules`: framing new every time; everyone around exactly like the images, no new crowds; each transition done in camera on the beat; no split screen. - `shots[]`: `cut_at` (on beats from `plan_music_montage`), `transition`, `camera`, `light`, `action`, `acting`, `head`, `performance` ("[raps/sings] every word in sync with Audio 1… nodding their head and shoulders on every beat"), `lyrics` (the words in that shot, from the word times). Picture roles only — never describe the person's looks. If the person cast another of their characters, name who performs the words in Audio 1. ## Shots and transitions - Two shots per frame: A — close to the frame's idea; B — another angle (dolly-in, low tracking, crane, over the shoulder, circling, an extreme close-up of eyes and lips, a Dutch angle, a push-in through the crowd). - Within one place: match on action, arc, crash cut, jump cut, roll, match motion. Between places: whip, object portal [an opening in the set, e.g. a porthole, a furnace mouth], wipe, shape, frame portal (a door), lap dissolve, flash cut, seamless (behind steam or a pillar), superimposition, smash cut. **No transition twice in a part.** - Write light per shot from the real sources of each place [furnace orange, a torch's blue-white, a red alarm, moonlight] and acting as eyes, lips, breath, a gesture. ## What to expect - Cuts land a little early (0.1–0.5 s); the edit trims each part to its length. `detect_video_cuts` shows where they fell. - Some transitions come out as plain cuts. - With a frontal face as Image 1 the head freezes; the five-angle board fixes it. ## Execute through MCP or on the board Use the person's FlashBoards account; never ask for a personal provider key or run the source's local scripts. Discover `get_account`, `list_generation_models` and the tools of this client first: a published guide does not prove a feature is deployed. **Headless, no open board.** Quote with `estimate_generation_cost` (skill models: `estimate_skill_model`), then run `generate_image`, `generate_video`, `generate_audio` or `run_skill_model` with one stable `generationAttemptId` per logical attempt. Pass `maxCostUsd` only when the current operation and selected model support it; otherwise use the quoted cost and the person's explicit approval with the supported controls. Never silently ignore an unsupported spending cap; make one `get_generation` check after `retryAfterSeconds`. Pass the person's own results by `generationId`: images in `referenceGenerationIds`, `firstFrameUrl` and `endFrameUrl`, audio in `referenceAudioUrls` or `targetAudioUrl`, videos in `referenceVideoGenerationIds`. They are read on the server; signed result links do not work as public URLs. Every `generationId` a tool returns stays valid: a later read of the same result gives a different string, and both work. Film frames and gallery pictures go in by `previewHandle`: `referencePreviewHandles` of the generation tools, a `{previewHandle}` source of `arrange_images`, `crop_image` and `pad_image`, and `describe_media`. New files come from `show_media_upload_picker` as receipts. The person's character comes in as its sheet, by `generationId` or as an upload; there is no search by name. If the person names a character without the sheet, ask for it or make one with `character-sheet`; do not scan recent results for it. Free server steps: `trim_video`, `concat_videos`, `conform_video`, `replace_video_audio`, `extract_video_frame`, `extract_video_audio`, `detect_video_cuts`, `crop_image`, `pad_image`, `arrange_images`, `cut_audio_span`, `concat_audio_tracks`, `mix_audio_tracks`, `fit_audio_duration`, `add_audio_reverb`. Check every result with `describe_media` (a question about an image; `model`, `mode` and `sourceRange` only for a video) and `transcribe_media` for speech. Both checks are paid: `describe_media` costs a fraction of a cent for an image and about 2–4 ¢ for a short video (more for a long one), `transcribe_media` under 1 ¢ a minute — count the checks you plan in the price you name before the first paid step. `describe_media` is not reliable about sound: it misses effects mixed under music and can hear voices or instruments that are not there. Check speech with `transcribe_media` and ask the person to listen for effects and music. Never repay an uncertain submission; reuse an attempt ID only for the identical request. Make each new `requestId` or `generationAttemptId` unique, for example the task, the date and a short random suffix: an ID used before, even in another chat, replays that earlier request or answers `REQUEST_ID_CONFLICT`. **On the board.** Inspect with `list_board_sessions` and `get_board_snapshot` and carry the same `sessionId` and a fresh `activeBoardId` as `expectedBoardId`. Native cards: `execute_board_action` (`listNativeModels`, `setModel`), `prepare_board_generation` with ordered board items or owned outputs, then one `generate_on_board` within the authorized budget. Board media tools: `list_media_tools` → `start_media_tool` → `get_media_tool_run`. Put headless results on a board with `add_media_to_board`. Keep mandatory safety settings and the live limits of each model. Prices, timings, endpoints and "free retry" claims in the sources are historical, not FlashBoards contracts. If a step this guide needs is not available or is refused (a tool error, `audio_not_configured`, a missing model or input), keep the finished assets, tell the person which step is blocked and why, and ask how to go on. A single generation that ends `failed` is not a blocked step: submit it once more with a new `generationAttemptId` (reworded if it may have hit the content checker), and stop to ask only if it fails again. A failed attempt is not always refunded, so the retry may be paid again and `maxCostUsd` caps each request on its own: count both against the budget the person approved. Never swap in another route on your own: no image model redrawing a layout or a frame, no `describe_media` in place of the frames it should check, no clip generated only to take its sound, no picture drawn where the guide takes a film frame. ## Source and related guides - [Original Learn guide](https://flashboards-h01.vercel.app/learn/formats-and-exhibition/ai-music-video/skills/musicvideo-shots.md) - Updated from the owner's skill sources of 2026-10-11. Read: `musicvideo`, `musicvideo-track`, `musicvideo-edit`. Source experiments and demo materials are historical; use the person's own approved media.