Claude Skill MartBrowse skillsQuick linesLearn by videoTerminal guideWhat is a Skill?
 
← Back to list

Image Prompting (Nano Banana & GPT Image 2)

Writes ready-to-copy image prompts for Nano Banana (NBP/NB2) and GPT Image 2, with model, quality and size recommendations.

Image & VideoIntermediate★ 165⑂ 23AI score 8/10Last updated: Aug 8, 2026

What it does

  • Routes the request first: picks Nano Banana (NB2/NBP) or GPT Image 2, since their prompt syntax differs fundamentally.
  • Writes in the right dialect — GPT Image 2 gets the 5-slot template (Scene / Subject / Important Details / Use Case / Constraints), Nano Banana gets natural prose.
  • Always returns a fixed Model / Quality / Size·Ratio / Prompt / Notes block you can copy straight into the tool.
  • For edits, emits an explicit Change / Preserve / Constraints list so faces, pose, lighting and framing don't drift.
  • Loads extra reference files only when relevant: text rendering, character continuity, slides, storyboards, multi-panel grids, plus vertical pattern libraries (e-commerce, fashion editorial, food & beverage, cinematic portraits, posters, UI/social).
  • Strips tag soup ("4k, epic, masterpiece") and negative phrasing, replacing them with hex colors, concrete materials and real lighting vocabulary.

Who it's for

  • Marketers and designers producing blog covers, ad visuals and product shots regularly.
  • Anyone whose AI image quality is inconsistent and wants a repeatable prompting method.
  • Creators building storyboards/keyframes that then feed an image-to-video workflow.

Examples

  1. "Write an e-commerce product shot prompt for this tumbler" → model choice + e-commerce pattern + 1536×1024 recommendation.
  2. Attach a reference image: "match this mood for our poster" → the skill decomposes lighting, grade and texture, then writes a style-transfer prompt.
  3. "Remove the background signage, keep everything else" → edit prompt with a preserve list for face, pose and lighting.

· · · Install guide · · ·

Try it now, no install

Paste this into Claude to use the skill without installing anything.

Read the instructions in this file and follow them to help me:
https://raw.githubusercontent.com/smixs/visual-skills/HEAD/image/SKILL.md

What I want: (describe your task here)

If Claude can't open the link, open it yourself and paste the contents instead.

↓ If it works for you, download the ZIP below and install it. Then it runs on its own — no pasting each time.

Install in the Claude app (no terminal)
  1. Download the ZIP with the button below.
  2. In Claude, open Settings → Capabilities and turn on 'Code execution and file creation'. (one time)
  3. Go to Customize → Skills → + → 'Upload a skill' and upload the ZIP.
↓ Download ZIP
Install in Claude Code

Let Claude do it — paste this into Claude Code

Install the skill I found on Claude Skill Mart.
Copy the image folder from the GitHub repo smixs/visual-skills into my ~/.claude/skills/image/.
When it's done, tell me in one line what this skill can do.

Install with a command instead

git clone https://github.com/smixs/visual-skills.git && mkdir -p ~/.claude/skills && cp -r visual-skills/image ~/.claude/skills/image

⚠ This is a third-party skill. Check the source repository before installing.

  1. Open a terminal.
  2. Clone the repo: git clone https://github.com/smixs/visual-skills.git
  3. Create the skills folder: mkdir -p ~/.claude/skills
  4. Copy the whole skill: cp -r visual-skills/image ~/.claude/skills/image — the references/ subfolder must come along, since SKILL.md is intentionally thin.
  5. Verify: ls ~/.claude/skills/image/references should list the per-model and per-vertical markdown files.
  6. Restart Claude Code, then try: "write an image prompt for a blog cover about coffee".
  7. Licensed CC BY 4.0 — credit the author (Serge Shima) when you publish outputs.
View source on GitHub ↗License: CC-BY-4.0