Skip to content

feiskyer/claude-code-settings / skills/gpt-image-skill

gpt-image-skill

Unverified

Generate or edit images using OpenAI GPT Image API (gpt-image-2, gpt-image-1, etc). Use ONLY when the user explicitly names OpenAI or GPT as the provider: "gpt image", "openai image", "generate image with openai", "用 openai 画图", "用 GPT 生成图片". For generic image requests without a provider, use nanobanana-skill instead. Do NOT use for diagrams (架构图/流程图) — draw those with Mermaid or code.

GPT Image Skill

Generate or edit images using OpenAI's GPT Image models through a bundled Python script.

Requirements

  1. OPENAI_API_KEY: Must be configured in ~/.gpt-image.env or export OPENAI_API_KEY=<your-key>
  2. OPENAI_API_BASE (optional): Custom API base URL for compatible endpoints (e.g. Azure OpenAI, proxies). Set in ~/.gpt-image.env or export it.
  3. Python3 with dependencies: openai, Pillow. Install via python3 -m pip install -r ${CLAUDE_SKILL_DIR}/requirements.txt if not installed yet.
  4. Executable: ${CLAUDE_SKILL_DIR}/gpt_image.py

Instructions

For image generation

  1. Ask the user for:

    • What they want to create (the prompt)
    • Desired size (optional, defaults to 1024x1024)
    • Output filename (optional, auto-generates UUID-based name if not specified)
    • Model preference (optional, defaults to gpt-image-2)
    • Quality (optional, defaults to auto)
    • Number of images (optional, defaults to 1)
  2. Run the script:

    python3 ${CLAUDE_SKILL_DIR}/gpt_image.py --prompt "description of image" --output "filename.png"
    
  3. Show the user the saved image path when complete.

For image editing

  1. Ask the user for:

    • Input image file(s) to edit (up to 3)
    • What changes they want (the prompt)
    • Output filename (optional)
  2. Run with input images:

    python3 ${CLAUDE_SKILL_DIR}/gpt_image.py edit --prompt "editing instructions" --input image1.png image2.png --output "edited.png"
    

Available Options

Models (--model)

  • gpt-image-2 (default) — Latest model with strong instruction following, text rendering, and broad world knowledge
  • gpt-image-1.5 — Mid-tier model
  • gpt-image-1 — First-generation GPT image model
  • gpt-image-1-mini — Lightweight, faster generation

Sizes (--size)

  • 1024x1024 (default) — Square
  • 1024x1536 — Portrait (2:3)
  • 1536x1024 — Landscape (3:2)
  • auto — Let the model decide

Quality (--quality)

  • auto (default) — Model decides optimal quality
  • high — Higher detail, slower
  • medium — Balanced
  • low — Fastest

Output Format (--format)

  • png (default) — Lossless
  • jpeg — Smaller file size
  • webp — Modern format, good compression

Background (--background)

  • auto (default) — Model decides
  • transparent — Transparent background (png/webp only)
  • opaque — Solid background

Other Options

  • --n <count> — Number of images to generate (default: 1)
  • --output <filename> — Output filename (default: auto-generated)

Examples

Generate a simple image

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py --prompt "A serene mountain landscape at sunset with a lake"

Generate with specific size and output

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py \
  --prompt "Modern minimalist logo for a tech startup" \
  --size 1024x1024 \
  --quality high \
  --output "logo.png"

Generate landscape image

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py \
  --prompt "Futuristic cityscape with flying cars" \
  --size 1536x1024 \
  --output "cityscape.png"

Generate with transparent background

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py \
  --prompt "A cute cartoon cat mascot" \
  --background transparent \
  --format png \
  --output "mascot.png"

Generate multiple images

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py \
  --prompt "Abstract art in the style of Kandinsky" \
  --n 3 \
  --output "art.png"

Edit existing images

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py edit \
  --prompt "Add a rainbow in the sky" \
  --input photo.png \
  --output "photo-with-rainbow.png"

Combine multiple reference images

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py edit \
  --prompt "Create a gift basket containing all items shown" \
  --input item1.png item2.png item3.png \
  --output "gift-basket.png"

Use a different model

python3 ${CLAUDE_SKILL_DIR}/gpt_image.py \
  --prompt "Detailed portrait of a cat in watercolor style" \
  --model gpt-image-1 \
  --output "cat-portrait.png"

Error Handling

If the script fails:

  • Check that OPENAI_API_KEY is exported
  • If using a custom endpoint, verify OPENAI_API_BASE is correct
  • Verify input image files exist and are readable (for editing)
  • Ensure the output directory is writable
  • Check that the model name is valid

Best Practices

  1. Be descriptive in prompts — include style, mood, colors, composition details
  2. For logos/icons, use square size (1024x1024) with transparent background
  3. For social media, use portrait (1024x1536) for stories or square for posts
  4. For wallpapers/headers, use landscape (1536x1024)
  5. Use high quality for final output, auto for quick iterations
  6. GPT Image models excel at text rendering — include text in prompts when needed
  7. For editing, provide clear instructions about what to change and what to keep
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 463kUpdated Claude CodeCodexVerifiedInstall
discernment-nudgeanthropics/skillsAfter you give a substantive answer or draft that the user may act on — advice or recommendations, drafted artifacts such as goals, plans, pitches, proposals, or emails, estimates or projections, analysis or interpretation of data, factual claims they may rely on, or a multi-step argument — invoke this skill BEFORE finalizing your reply and then, if it applies, append 2-3 short follow-up questions, each tied to something specific in what you just produced, that help the user check key facts, probe the reasoning or assumptions, and notice missing context. Do this at most once per conversation. Skip it when the user asked a trivial how-to or simple lookup, wants a purely educational explanation, asked you only to format, convert, or assemble a file from content they provided, is writing code they will run, is doing creative writing or casual chat, or already asked you to double-check, cite, or review — the skill file explains these boundaries and the exact output format.Stars: 195kUpdated Claude CodeCodexVerifiedInstall
internal-commsanthropics/skillsA set of resources to help me write all kinds of internal communications, using the formats that my company likes to use. Claude should use this skill whenever asked to write some sort of internal communications (status reports, leadership updates, 3P updates, company newsletters, FAQs, incident reports, project updates, etc.).Stars: 195kUpdated Claude CodeCodexVerifiedInstall