Image

按原图尺寸和构图转绘白底 Y2K 低清像素涂鸦，五官与全图共用一致的粗糙线条，保留表情，配松散涂色和微型贴图，并按主体与风景分配结构线密度，也可编写提示词。用于鼠标手绘、笨拙粗糙的像素画法。

Recreate a look from a reference image or node-editor screenshots in Blender, then measure the match - palette, exposure in stops, contrast, tint, focus, structure - and get numeric corrections. Includes safe workspace begin/end and a headless node-tree reader. Use when asked to copy, match or replicate an image or shader/Geometry Nodes screenshot.

Generate and edit images with Qwen-Image-2.0 (Alibaba, Feb 2026; one model for text-to-image and editing, native 2K, strong Chinese/English text rendering) through fal.ai (`fal-ai/qwen-image-2/...`) or Alibaba Model Studio (`qwen-image-2.0`, `qwen-image-2.0-pro`). Covers the availability check, tier and endpoint choice, explicit 2K sizes, prompts written by the official Qwen rewriting rules, exact quoted text for posters/slides/infographics/comics/calendars, edit instructions with image 1/2/3 re

Model-agnostic rules for generating and editing images of people and scenes with any image model (Qwen-Image, Nano Banana / Gemini Image, GPT Image, Seedream, Flux / Kontext, Kling O3 image, Midjourney, Stable Diffusion and newer ones). Covers intact human anatomy for every person (one head, two eyes, two arms, two legs, five fingers per hand, no limbs from the back or chest, no fused bodies), strict fidelity to attached references (inventory of people, objects and setting, change only what was

Build photoreal Blender assets from reference photos with one builder and an independent photo critic. Optionally prepare baked textures, LODs and collision for Godot or Unreal Engine. Use for photoreal props, scenes and reference-driven refinement in Blender. Start another refinement round only when the user requests it.

Use when making a conference, meeting, or hallway poster; turning a paper, preprint, or abstract into a poster; or revising poster layout, type size, and claims. Any field. No institution, logo, or topic is assumed. Triggers include poster, scientific poster, research poster, paper-to-poster, poster draft, and poster layout.

Analyze arbitrary text or structured content, plan visuals, and generate images using the standalone image-generation agent.

Restyle a supplied photograph as selective transit motion / 车流静止主体. Use for this specific visual method while preserving observed subjects and structure.

Restyle a supplied photograph as staggered photo ribbons / 错层照片带合影. Use for this specific visual method while preserving observed subjects and structure.

Generate anime-style images (and short anime videos) with AnimeAPI at $0.003 per image. Use when the user asks for an anime illustration, character art, avatar, waifu, manga-style scene, visual-novel sprite or background, or an anime video clip, and an ANIMEAPI_KEY is available. Runs a small bundled script that returns the image URL and saves the file.

Erstellt eine Schautafel: eine erklärende Übersichtsgrafik zu einem oder mehreren Themen mit Banner-Headline, Kacheln und Fußformel, in einem von zwölf Stilen (neo, classic, sepia, atlas, photo, sketch, toy, coach, glass, origami, lotus, messing) und frei wählbarer Kachelmatrix. Plant die Tafel als Gliederung, schreibt ein Prompt-Memo und lässt das Bild über den Skill image-generation von Codex erzeugen, in @Image/ ablegen und prüfen. Verwenden bei „Mach eine Schautafel zu …", „Schautafel zu …",

Restyle a supplied photograph as chisel carved wood animal / 刀刻木雕动物. Use for this specific visual method while preserving observed subjects and structure.

Restyle a supplied photograph as photocopy night splice / 复印拼缝夜景. Use for this specific visual method while preserving observed subjects and structure.

把人物设定或参考图转为有真人摄影质感的 AI 人像。适用于写实角色定妆、头肩照、面部特写，以及修正磨皮、假眼神、假发感或光影失真的现有角色图。可交付完整提示词，或在用户要求出图时生成并按成图验收。

Use when a user requests image generation, image editing, reference-based artwork, image styles, or multi-step image techniques through NodeClaw or a custom Images API, including tiled Pixel Art refinement.

Restyle a supplied photograph as hand tinted photo memory / 手敷淡彩照片. Use for this specific visual method while preserving observed subjects and structure.

Batch resize images and apply transparent PNG watermarks for blog posts and photo folders. Use this skill whenever the user asks to resize photos, add a watermark/signature PNG to images, batch-process a directory of blog photos, or launch the Watermark Resizer GUI.

Apply a consistent light-and-dark PowerPoint theme, including title slides, section dividers, closing slides, and optional miniature diorama artwork. Use for deck-wide restyling or section backgrounds. 明暗デザイン、表紙、区切り、箱庭背景の統一に使用。

Generates Mermaid diagrams of an existing codebase, written to a Markdown file that renders on GitHub. Produces an architecture overview (modules, services, data stores, external systems, and how they depend on each other) and, when the project has a schema, migrations, or ORM models, an ER diagram of the data model. Every box and arrow is backed by code. Use whenever the user wants to diagram, map, visualize, or document a repo's architecture, structure, or database, asks for a Mermaid or architecture diagram, or wants an onboarding overview of how a codebase fits together.

Help the user understand the current topic visually with concise diagrams, code-shape sketches, API-surface diffs, and focused HTML artifacts. Use when: explaining a topic visually, showing control flow, sketching component or file structure, comparing states, or creating a focused visual artifact. Triggers: show me, visualize this, draw the flow, diagram this, show the structure, API diff.

Plans, then builds, an automation that turns the photos a client has already sent into new static ad images with Gemini, using real reference photos so the scenes look real. Looks at one example client's photos in Google Drive, agrees what good variations look like for each kind of photo, tests a small batch, then sets it up to run on a schedule. Use when an agency wants more statics without asking clients for more photos.

Optimize an existing PowerPoint slide diagram around its intended message, using the slide text and supplied evidence while preserving the deck design. Use to clarify a diagram, adapt it to a topic, or choose a better visual structure. 既存の記述と伝えたい結論に合わせた図解の最適化、別トピックへの更新に使用。

Convert a photographed handwritten document (assignment, notebook pages) into a clean, complete document written in ONE chosen reference hand — matching the handwriting style, stroke thickness, ink colour and paper tone of a single approved reference page, page by page, with verified text accuracy. Use when asked to rewrite a document "in this handwriting", clone a hand across many pages, or turn photos of handwritten pages into a uniform printable PDF.

Professional logo and visual brand identity design: brand context capture, visual identity briefs, AI prompt engineering (DALL-E/Midjourney/FLUX/SDXL), SVG logo creation, raster-to-vector conversion (Pixel2Motion), logo animation (Disney 12 principles, CSS choreography), brand system documentation, font selection and sourcing, trademark awareness, and brand consistency standards. Use when asked to design a logo, create a brand identity, animate a logo, vectorize a raster logo, build a brand system, create a font pairing, or perform a trademark check. \" argument-hint: \"[brand name, style hints, or redesign request]\" allowed-tools: Bash, Read, Write, Edit, Glob, Grep, WebSearch, WebExtract, ImageGenerate, AskUserQuestion, TaskCreate, TaskUpdate