punk-avatar

name

punk-avatar

description

Generate avatar images and reusable avatar image prompts from the shared Punk style library for people, pets, objects, profile pictures, pet portraits, and avatar-derived keepsake cards. Use this skill to identify the subject from image or text, choose one avatar-capable style, compile the selected style atom into the avatar blueprint, save the final prompt, and generate the image when an image-generation tool is available.

Punk Avatar

Core Rule

punk-avatar is an avatar-shape workflow, not a style owner. Select exactly one reusable style from the repository-level styles/ library, then compile that style atom into the avatar shape.

The final image prompt is not a loose concatenation of "avatar instructions + style instructions". It must be one integrated, avatar-specific prompt that applies the selected style to a profile image, portrait avatar, object avatar, pet avatar, or avatar-derived keepsake card.

Use three inputs:

1. punk-avatar task fields: subject type, source image or text description, intended use, likeness policy, avatar composition, aspect ratio, background simplification, crop safety, and universal output constraints.
2. The selected style's META.md metadata and STYLE.md reusable visual style atom.
3. references/avatar-prompt-blueprint.md, which defines the full avatar prompt shape.

The style file defines the reusable visual language. The avatar blueprint defines the output form. The final prompt must fuse them into one complete avatar-generation prompt.

Resources

• Read references/style-catalog.md to list or choose avatar styles.
• Read references/avatar-prompt-blueprint.md before composing the final prompt.
• For the selected style, read both:
  • ../../styles/{style-id}/META.md
  • ../../styles/{style-id}/STYLE.md
• Expose only these six styles in the punk-avatar menu:
  • pixel-avatar
  • grotesque-soul-sketch
  • messy-crayon-pet-portrait
  • fashion-sketch-observation
  • polaroid-keepsake
  • minimal-paper-acrylic-block-illustration
• Do not expose cover/poster styles in the punk-avatar menu.
• Do not read punk-cover references or templates for avatar runs.

Workflow

1. Analyze the source material:

   • If the user provides one or more images, treat image input as primary.
   • Identify the main subject type: person, pet, object, or unclear.
   • Extract only concise visual traits needed for an avatar: face/head shape, hair or fur, eyes, expression, posture, signature colors, clothing, accessories, object silhouette, and any must-preserve marks.
   • Use text as supplemental context: name, temperament, intended use, preferred background color, elements to preserve, and elements to avoid.
   • If there is no image and the user describes a fictional avatar, handle it as description-based generation. Do not promise likeness to a real photo.

2. Resolve aspect ratio:

   • Default ratio is always 1:1 inside punk-avatar.
   • If the user specifies 3:4, 4:5, 16:9, or any custom ratio, keep the user's ratio exactly.
   • Do not use the selected style metadata's default_ratio as the avatar default.
   • For polaroid-keepsake, default to 1:1 in punk-avatar; only generate a vertical polaroid card when the user explicitly asks for a vertical ratio.
   • Do not ask a ratio question when the user omits ratio.

3. Confirm style before generating any prompt:

   • If the user specifies one catalog style, use it.
   • If the user supplies a complete visual direction that matches a catalog style, including the 极简纸感丙烯色块插画 brief, treat that style as specified and use its style atom without asking the user to repeat the style.
   • If no style is specified, recommend 2-3 eligible styles based on the subject type and ask the user to choose one.
   • When style is missing, stop after asking. Do not fill a style file, save prompt files, or generate an image.
   • Only auto-select one style when the user explicitly says to decide everything automatically.

4. Recommendation rules:

   • Pet subject: recommend 凌乱蜡笔宠物肖像, 拍立得纪念卡, and 怪诞灵魂手绘; include 像素头像 when a more icon-like avatar may fit.
   • Person subject: recommend 怪诞灵魂手绘, 时尚速写观察页, and 像素头像.

First Response Format

For input with no style, ask concisely:

默认画幅：1:1。

请选择一个头像风格：
- `风格 A`：reason tied to the subject.
- `风格 B`：reason tied to the subject.
- `风格 C`：reason tied to the subject.

也可以回复 "auto" 让我自动选择。

Do not ask for ratio unless the user explicitly wants ratio options.

Style Selection Heuristics

• Use pixel-avatar for people, pets, objects, symbolic avatars, icon-like profile pictures, and unclear subjects.
• Use grotesque-soul-sketch for people or pets that should feel expressive, funny, sketchy, and personality-driven.
• Use messy-crayon-pet-portrait only for pets and pet portrait avatars.
• Use fashion-sketch-observation only for people, especially fashion, travel, street-photo, film-still, or editorial personal portraits.
• Use polaroid-keepsake only for pets, especially memorial, collectible, named pet cards, and avatar-derived keepsake cards.
• Use minimal-paper-acrylic-block-illustration for people, pets, objects, scenes, or themes that need a simplified paper-texture illustration with vivid acrylic color blocks and a very small subject.

Avatar Prompt Blueprint

Use references/avatar-prompt-blueprint.md as the full prompt structure. The block below is retained only as the required task-field content that must be represented inside the integrated final prompt.

# punk-avatar task instructions

Create one single avatar image. Aspect ratio: {ratio}.

Use the following derived subject fields only:
- Subject type: {subject_type}
- Generation mode: {image_based_or_description_based}
- Subject identity: {subject_identity}
- Recognizable traits: {recognizable_traits}
- Name or text to include: {name_or_text}
- Intended use: {use_case}
- Background preference: {background_preference}
- Mood: {mood}
- Preserve: {preserve}
- Avoid: {avoid}

For image-based inputs, preserve the subject's most recognizable traits while translating them into the selected style. Do not claim exact biometric identity or photorealistic duplication.

For description-based inputs, create a fictional avatar from the description. Do not write that it preserves photo likeness.

The avatar must work as a profile image: readable at small size, centered or deliberately framed, clear silhouette, no important facial features or subject marks cropped out, simplified background, and one final output.

Generate only one final image. Do not output explanations, alternatives, grids, contact sheets, or multi-option compositions.

Output Discipline

• The final generated prompt must be one integrated avatar-specific prompt compiled from the avatar task fields, references/avatar-prompt-blueprint.md, and exactly one selected style atom.
• Do not paste the selected style STYLE.md as a standalone second section. Adapt its style atoms into the avatar blueprint.
• The final prompt must preserve the selected style's non-negotiable anchors from META.md and STYLE.md.
• Do not copy unrelated user text into the final prompt; use concise derived subject fields.
• Do not include style-selection rationale inside the prompt file.
• Do not combine multiple styles.
• Do not add cover/poster title hierarchy, article summary, or platform cover rules.