Video & Audio Production Skills

Designs and maintains semantic groupings of nodes on the filmmaking canvas — scenes, character-reference sets, act beats, and other reading-clarity frames. Use when nodes on the canvas cluster around a shared meaning (same scene, same character, same act, same production state) and would read more clearly with a titled frame around them. Don't force it — groups are a view concern, not an organizing tax.

Generates or edits images on the filmmaking canvas via the local generateimage.js CLI, following canvas-aware conventions for subtype stamping, reference wiring, and provenance edges. Use when the user asks to design a character, portrait, location, hero still, variation, edit, scene shot, multi-view reference sheet, or storyboard mosaic. CHARACTER DESIGN ROUTING (read first) — Whenever a character will appear in downstream video work (anything the user calls a video / clip / promo / 宣传片 / 短片 / 连续剧 / film / scene / 拍片 / shot — even if the canvas is empty and the character is being designed from scratch), DEFAULT to Pattern 7 (4-panel character reference sheet — Front/Profile/Back/Closeup), NOT Pattern 1 (single portrait). This holds whether the user has uploaded actor reference photos or not. With ≥3 actor refs, Pattern 7 uses multi-ref triangulation; with 0-2 refs, Pattern 7 uses text-only generation but still emits the 4-panel layout, because the video model needs multi-view anchor data to keep the character recognizable across non-front shots. Pattern 1 (single portrait) is only for one-off static stills that will NOT feed video gen — character posters, print art, standalone illustrations. Before any character generation, briefly announce the choice to the user ("Starting with a 4-panel reference sheet for each character so the identity stays locked across video shots — tell me if you want simple portraits instead"). Never silently default to Pattern 1 when video work is implied. STORYBOARD SHAPE — storyboard / mosaic / 分镜 / shot list / keyframe sheet / coverage / previs = ONE composite image with N×M panels per location, NOT N separate calls. Re-invoke this skill (Pattern 6) before firing the CLI.

Handles screenplay work on the filmmaking canvas, only on explicit user intent (never on a file drop). Triages input as an official screenplay (use verbatim), a story or concept (iterate then rewrite at the target duration), or neither (defer to image/video skills). Captures the finalized script as a canvas note and sets the workflow title. Separately, on an explicit user command, splits the script into shot notes capped at 15 seconds each and offers to design character portraits plus location stills. Use when the user asks to write, adapt, or rewrite a script, story, or screenplay; to split a script into shots or clips; to extract or design characters or locations from a script; or to analyze or break down a canvas script. Preserves dialogue verbatim. Does NOT split or analyze without explicit user intent.

Orchestrates story, script, screenplay, concept, product promo, and multi-shot idea work into finished video. Use first when the user asks to make a video from a story or script; asks what next in a story video project; or needs a decision spanning script splitting, image refs, voices or VO, video clips, render strategy, Timeline ordering, or final Timeline handoff. Routes execution to script-compose, image-compose, voice-compose, and video-compose before those skills' CLIs are used.

Generates short videos on the filmmaking canvas via the local generatevideo.js CLI, following canvas-aware conventions for prompt composition, reference attachment, and provenance edges. Use when the user asks to animate, film, shoot, or generate a video of a canvas character, scene, or image; continue or extend an existing clip; restyle, partially edit, replace, or re-plot an existing clip; attach a character voice as reference audio; borrow camera moves, action, or VFX from a reference clip; render a script or its shot notes as a video sequence; render a storyboard mosaic as a video; compose an ad, brand film, or music-video shot; or generate a standalone clip.

Designs and attaches voices to characters on the filmmaking canvas via the local generatevoice.js CLI, following canvas-aware conventions for voice description, sample text, and attachment to the target node. Use when the user asks to give a character a voice; design a voice, sample, or read; preview how a character sounds; or generate a standalone narration / voice-over track that isn't tied to a specific character.