raphaelbgr

dialogue-and-voice

Use when a generated clip needs someone to speak, when a clip came back with invented speech in the wrong language, when lip-sync cuts off mid-line, when unwanted text renders on screen, or when the audio mix sounds cluttered. The dialogue, voice-profile and audio-layer rules - including the quotation-mark question and the ellipsis bug that silently kills lip-sync.

identity-and-likeness

Use when a real or recurring person must appear in generated video - choosing between a consent-gated avatar, reference photos, reference-to-video, or a trained LoRA; writing prompts that preserve likeness instead of fighting it; and applying the consent and policy rules. Covers the zero-shot gate that tells you whether training is needed at all.

multiclip-continuity

Use before delivering ANY multi-clip AI video script, and after every round of prompt revisions - audits consecutive clip pairs for spatial, prop, wardrobe, cast, knowledge, time and payoff breaks, then produces a regeneration ledger saying exactly which clips to re-generate and which to keep. Model-agnostic; applies to Veo, Omni Flash, Seedance, Wan, Kling and anything else that renders one clip at a time.

prompt-expander

Use when someone hands you a rough video idea, logline or one-liner and wants a full cinematic prompt, when writing or polishing a Gemini Omni Flash / Google Flow prompt, or when planning a conversational edit of an existing clip. Turns "a lonely astronaut finds a flower" into a complete, generation-ready prompt by deciding every dimension a clip can express.

seedance-prompting

Use when planning, pricing or prompting ByteDance Seedance video generation (2.5 or 2.0-mini) - reference-to-video with character consistency, lip-synced dialogue, video extension/editing, or deciding whether a shot is worth generating at all. Encodes the billing formula, the per-second cost tables, and the block-cheap / finish-expensive workflow.

veo-flow-prompting

Use when writing or fixing prompts for Google Flow video generation - Veo 3.1 or Gemini Omni Flash - including multi-clip scripts, dialogue clips, silent clips, the @me avatar, on-screen text, and clips that keep getting blocked by the content filter. Encodes the verified prompt anatomy, the per-clip complexity budget, and the failure modes that cost real credits.

Code & Development Listed

veo-prompt-lint

Use before submitting a Veo prompt or a batch of clip prompts, and when a generated clip came back with unwanted speech, the wrong camera motion, ignored references, or a wrong environment after a cut. Runs a regex linter over prompt text and proposes rewrites for the gotchas that recur every episode. Triggers on "lint veo prompt", "check this veo prompt", "veo prompt review".

viral-short-script

Use when writing or planning the script and structure of a short vertical video (Reel/TikTok/Short, under ~60s) - a visual-first, sound-off scroll-stopper skeleton. Triggers on "viral script", "hook for a reel", "structure of the video", or feeding a video-generation pipeline. Theme-agnostic - it produces the skeleton; the topic is plugged in afterwards.

wan-3-prompting

Use when writing or fixing prompts for Wan 3.0 video generation (Alibaba), via any API route - multi-clip film scripts, reference-to-video with real people, or when generated scenes come out illogical (objects that refuse to disappear, actions that get skipped, the wrong person speaking). Encodes the official Alibaba prompt rules, the API surface that actually exists, and the failure modes already paid for.