stage-generate — скилл для Claude Code: генерация видео и установка

---
name: stage-generate
description: AI-generated footage plus execution of bounded semantic video-edit segments already approved by EDIT/AUTO. Generate per shot with `ovs video`, preserve signed reference/settings, then assemble. For recurring characters also read stage-consistency.
---

# stage-generate

How to produce AI-generated footage and execute a **bounded semantic video edit** already planned by EDIT/AUTO. Designed HTML remains composition work; deterministic cutting remains stage-edit work. The generation tools are `ovs image`, `ovs video`, and `ovs speak`, followed by `ovs edit` assembly. Every billable call belongs to a signed generate segment.

## Pattern A — talking-head / spokesperson

1. **Character still:** generate one image of the presenter / avatar with the intended look. **Keep this reference image** and reuse it for every shot of the same character.
2. **Bring it to life:** generate a video *from* that image (image-to-video). When the provider returns speech + **built-in audio**, that audio is the deliverable voice — it is **lip-synced to the mouth in the clip** — so keep it and do NOT synthesize a separate narration. Only when the clip comes back **silent** do you synthesize the narration (`ovs speak`) and add it as the audio track. Synthesizing a fresh TTS track over a clip that already speaks is the #1 talking-head defect: the new audio has different wording/timing/length, so the voice no longer matches the lips.
3. **Polish:** add captions / a lower-third / a hook by authoring a small composition (stage-compose) and overlaying it onto the clip — **visual-only**. Preserve the clip's own (lip-synced) audio through assembly; a captions composition must not carry a narration `<audio>` track that would replace the clip's voice.

## Pattern B — cinematic / AI b-roll montage

1. **Storyboard** the shots (each: prompt, camera motion, duration).
2. **Generate each shot** (one `ovs video` call per shot; reuse a shared reference image / consistent style prompt for visual continuity). Pass the Gate-C-approved `--image-urls`, `--ratio`, `--duration`, `--resolution`, and `--generate-audio` values exactly; do not let provider defaults silently replace the signed plan.
3. **Assemble:** concatenate the shots in order, add transitions, and overlay a title / captions from a composition.

## Consistency (basic — deep consistency is a separate skill)

- **Within a shot:** drive the clip from a reference image (image-to-video) to lock the subject.
- **Across shots:** reuse the **same** reference image / style prompt.
- Full multi-shot character consistency, Cameo (upload-a-photo-as-the-lead), and long-narrative planning live in `stage-consistency`.

## Director judgment (generation line)

Craft calls specific to AI-generated footage, on top of the shared craft reference (video-craft):

**Talking-head / spokesperson**
- Understand what's said before placing overlays; time graphics to the spoken words.
- **3–6 overlays/min**, varied types; keep them in speaker-safe zones — never over the face.
- Cut silences and filler; for vertical, keep subtitles low so they don't cover the face.

**Cinematic / AI b-roll**
- Open on a hero frame; keep a small transition palette (cut / fade-to-black / slow dissolve / restrained push-in).
- Protect earned moments — don't over-cut a held look or a deliberate silence.
- Design each shot first-frame → last-frame and let audio dynamics carry momentum (shot language: video-craft §10; identity across shots: stage-consistency).
- **Frames are static snapshots, never an action in progress** (`video-craft` §10); in motion / last-frame text, name characters by visible features, not names (the model conditions on pixels, not labels).
- **Spend keyframes by how much the shot changes (cost gate).** If a shot's start and end look nearly the same — a talking head, a small pose/expression change, a gentle pan — it needs only ONE keyframe and motion fills the rest (variation_type `small`). Only a shot that ends somewhere visually different — a new subject enters, a wide→close transition, a big camera move — needs TWO keyframes for the model to interpolate (`medium` / `large`). Don't pay to generate an end-frame you don't need.

## Default scope caps (cost control — do not exceed without explicit user request)

Generated clips and images are **billable hosted calls**, so bound the run by default:

- **Shots / clips: ≤ 6** per video.
- **Characters: ≤ 3** per video.
- **One aspect ratio** per run.

If the brief seems to need more (a long story, many scenes, many characters), DO NOT silently fan out — state the larger count + the rough number of billable generations in the pre-generation confirmation and let the user opt in first. Treat anything above these caps as requiring explicit confirmation.

## Rules

- **Cost/time discipline:** every generated clip is a hosted, billable, multi-second call. State the exact shot/character count in the pre-generation confirmation; never start generating before the user has approved the count.
- **Audio (talking-head):** a generated talking-head clip's built-in audio is **lip-synced** to its mouth. Treat it as the final voice: keep it through assembly and finalize. NEVER add or mux a separately-synthesized narration over a clip that already speaks — it desyncs from the lips. Synthesize narration ONLY for a silent clip, or for b-roll / off-screen voiceover where no mouth is visible.
- **Brief drives params:** record and pass the exact reference images, aspect ratio, per-shot duration, resolution, and audio choice to each generation call.
- **One authorization, one output set:** every billable segment in the approved plan must have a matching result. A retry that creates a new hosted task or output set needs fresh Gate-C authority; recovery may inspect or download an already-authorized result without asking again. Use `ovs gate transition` instead of inventing line-local confirmation rules.
- **Semantic edit boundary:** `operation:"edit"` requires a top-level video reference with `intent:"edit"`, target/temporal anchors, plus `edit_strategy.mode:"semantic"|"mixed"`. The prompt may change only `may_change`; protected identity/content/motion/timing/audio axes remain explicit. Never turn an edit into unconstrained regeneration.
- **Partial batches:** preserve completed sibling segments. A revision or failure invalidates only that shot plus downstream assembly evidence; never redispatch completed shots merely to rebuild the montage.

## Boundary / non-goals

Generation-primary work uses this skill directly. EDIT/AUTO may invoke it only as executor for an approved semantic-edit segment; source analysis, preservation boundaries, and final assembly stay with the edit plan. Designed HTML goes to stage-compose and deterministic cuts/joins to stage-edit.