Stats
Actions
Tags
From Nim
Use when generating believable editorial young-adult human portraits or fashion/UGC character images with Nim, especially Gen-Z fashion casting, model/person look development, aesthetic human references, or prompt-driven human generation. Asks for age or age range, sex/gender presentation, ethnicity/look, and additional notes; defaults to Nano Banana Pro in 9:16, but supports user-specified models such as GPT Image 2 by discovering the live Nim model contract.
How this skill is triggered — by the user, by Claude, or both
Slash command
/nim:nim-person-generatorThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
Generate aesthetic, believable editorial young-adult humans through the Nim MCP. The goal is a castable, specific, modern person with distinctive grooming, posture, attitude, and subtle subculture styling.
Generate aesthetic, believable editorial young-adult humans through the Nim MCP. The goal is a castable, specific, modern person with distinctive grooming, posture, attitude, and subtle subculture styling.
Ask only for missing details that materially affect the person:
If the user says to decide, make tasteful assumptions and state them briefly before generation. Do not ask for every optional styling detail.
age 21-25 by default.Use Nim image generation.
Nano Banana Pro / text-to-image.9:16.2K, unless the user asks otherwise.1. Use 2-4 only when the user asks for options.Always discover the current model with models_explore and inspect the exact model contract with models_explore action=get before generation. Pass only parameters allowed by the returned generationContract.
If the user requests another model, search for that model by name, get its contract, and adapt the allowed parameters. Examples: gpt image 2, Seedream, Flux, or another Nim catalog model.
finished, failed, or cancelled.Do not dump raw JSON, internal IDs, or long prompt mechanics unless the user asks.
Use this source prompt as the core style system:
Create an aesthetic, believable, editorial young adult for a Gen-Z fashion. The person should feel cast, specific, and modern, with distinctive grooming, posture, attitude, and subtle subculture styling. Do not create an ugly person; balance the appearance parameters finely.
natural skin texture, subtle film grain, shallow depth of field, cinematic realism like a 50mm lens at f/2.0.
COLOR GRADE / LUT STYLE: TikTok Beauty Soft LUT, soft lifted shadows, low contrast, pastel warmth, pink-beige undertone, clean bright face exposure, soft whites, natural lips and cheeks, flattering modern beauty UGC. Add 50D ultra-fine film grain at very low opacity, minimal halation on glossy product edges, soft cosmetic highlight glow. Camera enhancers: stabilized creator camera, clean eye autofocus, slight exposure breathing, soft lens diffusion, realistic compression texture, preserved skin pores. No plastic skin, no face smoothing artifacts, no over-sharpened eyes.
Build concise prompts with this structure:
Vertical 9:16 editorial portrait / fashion UGC image.
PERSON: age <age or range>, <sex/gender presentation>, <ethnicity/look>. Specific castable face, modern Gen-Z styling, distinctive grooming, posture, attitude, and subtle subculture cues.
WARDROBE / STYLING: <user notes or tasteful assumption>.
SCENE: <location/background if specified or a simple editorial setting>.
CAMERA: cinematic realism, 50mm lens at f/2.0, shallow depth of field, clean eye autofocus, stabilized creator camera feel.
LIGHTING: flattering bright face exposure, soft whites, pastel warmth, pink-beige undertone, low contrast, soft lifted shadows.
TEXTURE: natural skin texture, preserved pores, subtle 50D ultra-fine film grain, realistic compression texture, soft lens diffusion, minimal halation on glossy edges.
MOOD: believable, specific, aesthetic, modern, editorial, not generic.
CONSTRAINTS: no plastic skin, no face smoothing artifacts, no over-sharpened eyes, no distorted hands, no fake text, no watermark.
Use these only when the user leaves styling open:
When generating multiple options, vary one or two controlled axes only:
Keep age, sex/gender presentation, and ethnicity/look consistent unless the user asks to explore those.
Got it. I need four details before generating: age or age range, sex/gender presentation, ethnicity/look, and any extra styling notes. If you want me to decide, say "choose for me" and I will use tasteful Gen-Z editorial defaults.
Vertical 9:16 editorial portrait / fashion UGC image.
PERSON: age 22-25, female-presenting, mixed Mediterranean and Eastern European look. Specific castable face, modern Gen-Z styling, distinctive dark micro-bob haircut, minimal silver piercings, relaxed posture, self-possessed attitude, subtle art-school streetwear cues.
WARDROBE / STYLING: cropped charcoal knit, low-rise washed black denim, narrow belt, understated glossy lip, natural cheeks.
SCENE: soft daylight apartment wall with a slightly messy mirror edge and muted fashion-magazine realism.
CAMERA: cinematic realism, 50mm lens at f/2.0, shallow depth of field, clean eye autofocus, stabilized creator camera feel.
LIGHTING: flattering bright face exposure, soft whites, pastel warmth, pink-beige undertone, low contrast, soft lifted shadows.
TEXTURE: natural skin texture, preserved pores, subtle 50D ultra-fine film grain, realistic compression texture, soft lens diffusion, minimal halation on glossy edges.
MOOD: believable, specific, aesthetic, modern, editorial, not generic.
CONSTRAINTS: no plastic skin, no face smoothing artifacts, no over-sharpened eyes, no distorted hands, no fake text, no watermark.
Guides creation, editing, and verification of skills for AI coding agents using test-driven development with subagent scenarios. Use when authoring or debugging skills.
npx claudepluginhub nim-video/skills --plugin nim