prompts.

Key flags: - `-o/--output` output PNG path - `--color` `white|black|#RRGGBB` - `--tolerance` color matching tolerance - `--overwrite` overwrite original file Example:

\ -o photosynthesis-es.png \ --history-in photosynthesis-history.json \ --history-out photosynthesis-history.json ``` History files preserve the conversation contents needed for follow-up turns, including thought signatures returned by the API. ### `icon` Usage: ```bash nanobanana icon [prompt] -o OUTPUT ``` Key flags: - `-o/--output` output directory or naming pattern - `--sizes` comma-separated icon sizes - `--style` `modern|flat|minimal|detailed` - `--background` `transparent|white|black|#RRGGBB` Examples: ```bash nanobanana icon

-o ./icons/ --sizes 16,32,64,128 ``` ### `pattern` Usage: ```bash nanobanana pattern [prompt] -o OUTPUT ``` Key flags: - `-o/--output` output file path - `--size` tile size as `WxH` - `--style` `geometric|organic|abstract|floral|tech` - `--type` `seamless|texture|wallpaper` Examples: ```bash nanobanana pattern

First-person perspective, user interacts with holographic workspace in midair, camera captures floating interfaces responding to gestures, soft beeps and electronic chimes fill the air, cool ambient lighting, immersive futuristic cinematic realism style

Upload Flux image and selfie, add yourself to the background.

Change the facial expression to happy and make the weather sunny.

Generate better image of mom's 60th birthday party cake for bakery.

Transform the scene as per description.

Add hair to the bald man in the image.

pnpm db:push # Push schema changes (development) pnpm db:generate # Generate migrations pnpm db:migrate # Run migrations pnpm db:studio # Open Drizzle Studio GUI pnpm db:reset # Drop and recreate all tables

A cat walking in the rain, cinematic lighting.

Product demo of a smartwatch with rotating camera view.

Nature scene with flying birds and ocean waves sound.

beautiful stunning ultra-detailed 4K 8K masterpiece trending on artstation cinematic lighting professional photography premium quality

Create a pitch-deck slide titled "Q3 Revenue Performance" that looks like a real Series A board-meeting slide. Layout (16:9): title top-left, 36pt Inter dark gray. Two-column body: left 60% chart, right 40% three KPI cards. Chart: vertical bars, Q1–Q3 2026, y-axis $0–$8M, three bars at $3.2M, $4.8M, $6.5M, muted blue palette. KPI cards: "+34% YoY", "189 new accounts", "$42K ACV". White background, Inter typography, tight 8px grid, no clip art, no gradients, no stock photography.

注意里面**没有**:没有夸赞词、没有"ultra detailed"、没有"8K"。每一个词都在做具体的指令性工作。 Skill 教 Claude(或任何 agent)规范结构:

Skill 就是文本 + 一个脚本。无构建、无重启、无注册。 ### 给你的团队定制 Fork 一份。设计上可以替换的部分: - **`references/use-cases.md`** — 加你们自家风格、品牌色卡、常用素材模板(自家的 pitch-deck 模板、自家的品牌 hex、自家的信息图格式) - **`scripts/gpt_image.py`** — 加 `--watermark` flag、自定义输出命名、S3 上传、Slack 推送、随便加 - **`DEFAULT_BASE`** — 指向你们的私有网关而不是默认值 - **agent 拿到的 system prompt** — 如果你们的 agent 不会自动发现就把 `SKILL.md` 接进 bootstrap 通用的改动欢迎 PR。团队特定的留在你的 fork 里。 ### License MIT。看 `LICENSE`。随便用。 ### 致谢 为 [Claude Code](https://claude.com/claude-code) 而做,但这个模式(markdown skill + 零依赖 CLI + 视觉自验证)适用于任何能读文件 + 调脚本的 agent。 灵感:OpenAI 官方 GPT Image 2 Cookbook,Anthropic superpowers/writing-skills 关于 agent 可发现 skill 的模式。 --- ## English > Drop this into Claude Code or Codex. Your agent will go from "here's a curl command, good luck" to producing pitch-deck slides, Chinese posters, pixel-art tilesets, photoreal product shots, and surgical photo edits — first try, every try. GPT Image 2 dropped in April 2026. It's a generational leap: long instructional prompts no longer lose detail, text rendering is finally correct (Chinese / Japanese / Korean too), custom resolutions go up to 3840px, and the edit endpoint does precise local edits via a `change ONLY X / preserve Y exactly` pattern. But here's the catch: agents trained before April 2026 don't know any of this. They write prompts the old way — *"4K, ultra detailed, masterpiece, trending on artstation"* — which on GPT Image 2 is at best ignored and often actively harmful. They paste curl commands instead of saving files. They generate, then ask the user "does this look right?" instead of looking themselves. This skill fixes all of that. ### What you get

4K, ultra detailed, masterpiece, trending on artstation

Every panorama is one step in a walking route. - The camera stays at human eye height.

## 📸 Examples Here are some scientific figures generated using FigForge: ### Sample Input - Neural Architecture  ### [LiveSearchBench](https://arxiv.org/abs/2511.01409)  ### [ReSo](https://arxiv.org/abs/2503.02390)  ### [VIKI-R](https://arxiv.org/abs/2506.09049)  > All figures are generated with clean conference-style design, featuring flat aesthetics, consistent line weights, and professional color palettes. --- ## 📖 How It Works ### Step 1: MODULE LIST Generation (GPT-5) The GPT-5 model analyzes your scientific text and creates a structured MODULE LIST that breaks down your architecture into: 1. **Input(s)**: Data sources and preprocessing 2. **Preprocessing/Encoding/Embedding**: Feature extraction layers 3. **Core Architecture/Stages/Blocks**: Main model components in sequence 4. **Special Mechanisms**: Attention, memory, routing, etc. 5. **Output Head**: Final prediction layers ### Step 2: Figure Generation (Gemini-2.5-flash-image) Using the MODULE LIST as a guide, nano banana generates a clean, professional figure following these design principles: - ✅ Flat, clean conference style (no gradients, shadows) - ✅ Consistent thin line weights - ✅ Professional pastel color palette - ✅ Rounded rectangles for module blocks - ✅ Clear arrows indicating data flow - ✅ Concise labels (no long sentences) - ✅ Pure white background with clean spacing ## 📁 Project Structure

for MODULE LIST │ └── step2_figure_generation.txt # Prompt for figure generation ├── examples/ │ └── sample_input.txt # Sample scientific text ├── outputs/ # Generated figures (auto-created) └── README.md # This file

生成每张幻灯片图片 3. **版本管理**：每次生成结果作为新版本保存，支持切换与回溯 ### 主要 API - 演示文稿：

[NAME] = Your name [SUBJECT] = The person shown in the uploaded reference image Make a symbolic portrait reinterpretation for artistic editorial use, using a medium-close framing at eye level with a calm, centered composition, set within an abstract environment derived from the etymological and cultural meaning of [NAME], with [SUBJECT] placed as the sole figure in the midground, preserving the facial structure, facial proportions, expression, and overall likeness of [SUBJECT] exactly as in the reference image, allowing no alteration, stylization, or reinterpretation of the face under any circumstance, avoiding any reuse, reference, or derivation of clothing, accessories, or styling from the reference image, and instead designing a new outfit that supports the mood and symbolism of [NAME] without echoing the original wardrobe, translating the meaning of [NAME] into visual elements such as light behavior, color palette, atmosphere, natural forces, or metaphoric forms surrounding [SUBJECT], reflecting the name’s meaning through mood and symbolism rather than literal text, letters, or icons, using lighting direction, intensity, and color temperature to reinforce the emotional essence of [NAME], integrating symbolic elements so they feel physically and spatially connected to [SUBJECT] rather than decorative, keeping the background restrained and uncluttered with no readable text, and ensuring strong visual impact, coherence, and stylistic integrity throughout the composition.

Make a photorealistic studio head portrait of the person from the provided reference photo, used strictly as identity anchor, captured at eye level with natural facial perspective, framed as a clean studio photograph with only the head and upper shoulders visible, set against a professional studio background with soft neutral tones, showing the subject centered and facing forward with a calm direct gaze, with a single hand gently closing a metal zipper over the lips, ensuring the lips and zipper merge naturally and convincingly without distortion, rendered with realistic skin texture and fine detail, lit with soft controlled studio lighting to create an elegant, cinematic, and emotionally restrained image.