メニュー
Build consistent character designs and character media with `vg generate`. Use this for original characters, reference sheets, expression sheets, outfit variations, identity-preserving edits, and character-to-video workflows.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Use the `vg generate` CLI to search, inspect, run, and manage 1200+ generative model endpoints. Trigger when the user asks to "generate an image", "make a video", "search models", "run a model", "fetch schema", "check pricing", "upload an asset", "queue async job", "track request", or any direct interaction with the model endpoint catalog. This is the foundational skill. Every other media skill in this repo executes its work through `vg generate` commands. Use `--json` whenever the output will be parsed by an agent.
Build 3D browser apps AND games with Three.js (r150+, ES modules): scene setup, geometries, materials, lighting, animation, GLTF models, physics (Rapier/cannon-es/arcade), character controllers, follow/third-person cameras, fixed-timestep loops, post-processing, and performance/debugging. Use for 'create a three.js scene/app/showcase', any 3D web content, and turning a scene into a game — 'add physics to my three.js game', 'third-person/character controller', 'follow camera', 'jump and gravity', 'collide with / pick up objects', 'structure my 3d game'. Trigger: threejs, three.js, 3D scene, WebGL scene, GLTF/GLB, OrbitControls, Rapier, cannon-es, 3D game, third-person controller. For 2D games use `phaser`; for engine-agnostic feel/balance/level craft use the game-craft skills.
Turn a character anchor into an engine-loadable animated spritesheet by generating ONE labeled pose-board image (a grid of the same character in the frames of an action) and slicing it. Works for any action — idle, run, jump, attack, hurt, crouch, death, roll. Generates a per-frame-labeled pose board on a flat chroma matte, recovers/slices the frames, keys + despills, snaps to crisp native pixels, normalizes with headroom, and packs spritesheet.png + a manifest. Triggers: 'sprite animation', 'animated spritesheet', 'attack animation', 'walk/run cycle', 'animate this character', 'game sprite animation', 'sprite pose sheet'.
Interview the user relentlessly about the game they want to make until the idea is build-ready, resolving each branch of the design tree one question at a time. Use when the user has a fuzzy or one-line game idea and wants to flesh it out before building: 'grill me about my game', 'help me figure out what game to make', 'stress-test my game idea', 'interview me about the game', 'I want to make a game but I'm not sure what'. Not for when the user already knows what they want built — go straight to game-playbook.
Build and ship Three.js apps on Capacitor iOS with Vite and Swift Package Manager: GLTF loading, assets_index animation UI, OrbitControls mouse/touch mappings, and iOS sync/run troubleshooting.
Build 2D browser games with Phaser 4: scene lifecycle, sprites, Arcade/Matter physics, tilemaps, WebGL rendering, filters, lighting, shaders, DynamicTexture/RenderTexture, SpriteGPULayer, TilemapGPULayer, and Phaser 3 to 4 migration. Trigger: phaser, phaser 4, phaser v4, create phaser game, add phaser scene, phaser sprite, phaser physics, phaser tilemap, phaser arcade, phaser webgl renderer, phaser filters, phaser SpriteGPULayer, phaser TilemapGPULayer, migrate phaser 3 to 4.
| name | character-design |
| description | Build consistent character designs and character media with `vg generate`. Use this for original characters, reference sheets, expression sheets, outfit variations, identity-preserving edits, and character-to-video workflows. |
vg generateUse this skill when the user wants to create, refine, or preserve a character. Load the reference files when needed:
references/anchor-system.mdreferences/prompt-patterns.mdreferences/examples.mdLoad model-catalog alongside this skill for default endpoint choices.
The main objective is consistency. Keep the character anchor stable and change only the requested scene, expression, outfit, camera, or action.
Only ask for missing inputs that affect identity or model routing.
model-catalog defaults unless the user names one.Follow the standard workflow in the generate skill: resolve endpoint → inspect schema/pricing → upload references → run (stills inline, video --async + status) → download to ./outputs/characters/{request_id}_{index}.{ext}. Default endpoint IDs are in the Model routing section below; routing details live in model-catalog.
Create a short immutable anchor before generating.
CHARACTER ANCHOR:
[name or codename], [age range], [face shape], [eye shape and color],
[nose and lips], [skin tone and distinguishing marks], [hair color, texture,
style], [body build and posture], [signature clothing or silhouette],
[style target]
Then add a variable block for the current shot.
SHOT VARIABLE:
[expression], [pose/action], [outfit changes if allowed], [environment],
[camera/framing], [lighting], [mood]
Never rewrite the anchor casually. If a result changes identity, strengthen the anchor or switch to a reference/edit workflow instead of adding more style words.
openai/gpt-image-2.fal-ai/nano-banana-pro or
fal-ai/nano-banana-2.fal-ai/flux-2/klein/9b.openai/gpt-image-2 first;
if editing an existing image, inspect openai/gpt-image-2/edit.fal-ai/nano-banana-pro/edit,
then openai/gpt-image-2/edit, then
fal-ai/bytedance/seedream/v5/lite/edit.bytedance/seedance-2.0/image-to-video for final quality.xai/grok-imagine-video/image-to-video.veed/fabric-1.0,
veed/fabric-1.0/text, or fal-ai/creatify/aurora.Reject or retry when:
Return downloaded paths and include the anchor used so future prompts can reuse the same identity.