Skip to content
mcp/skillhub

Agent Skill

Get Prompt From Image skill

get-prompt-from-imagewuyoscar/GPT-Image2-Skillskills/get-prompt-from-image/SKILL.md

Analyze user-provided reference images and reverse-engineer high-fidelity AI image-generation prompts. Use when the user asks to recreate, imitate, reverse-engineer, or extract prompts from photographs, illustrations, 3D renders, products, characters, landscapes, typography, logos, posters, or other visual references. Do not use for requests that only require OCR or an ordinary image description.

5.8kMIT

always on

83 tokname + description

once used

1,373 tokSKILL.md body

bundled files

420 KB

repo stars

5.8kGitHub

Install the Get Prompt From Image skill

Skills CLI (Claude Code, Codex, Cursor and others)

npx skills add https://github.com/wuyoscar/GPT-Image2-Skill --skill get-prompt-from-image

Claude Code, by hand (all projects)

git clone --depth 1 https://github.com/wuyoscar/GPT-Image2-Skill.git /tmp/GPT-Image2-Skill
mkdir -p ~/.claude/skills
cp -R /tmp/GPT-Image2-Skill/skills/get-prompt-from-image ~/.claude/skills/get-prompt-from-image

For one project only, copy the folder into .claude/skills/ in that project instead. The agent picks it up when a task matches its description.

What it costs your context

Agent Skills load in three steps. Tokens counted with the o200k tokenizer; your model's count will differ a little.

WhenWhat loadsCost
Every sessionName and description83 tok
When the agent uses itThe SKILL.md instructions1,373 tok
Only if opened4 bundled files (20 KB)varies

Until it is used, get-prompt-from-image costs about 16× less than the median MCP server we measured (1,362 tokens on every turn). Installing all 2 skills from wuyoscar/GPT-Image2-Skill would load 145 tokens of descriptions. Skills vs MCP study

Checks

Our checks found nothing to flag: it follows the Agent Skills specification's naming and size guidelines and doesn't pipe downloads into a shell.

Checked 2026-10-22 · method

What's inside

The headings of its SKILL.md. Read the full file on GitHub.

  • Get Prompt from Image
  • Core Principles
  • Workflow
  • Medium Boundaries
  • IP, Brands, Logos, and Text
  • Default Output Format
  • 1. Positive Prompt
  • 2. Negative Prompt

More skills from wuyoscar/GPT-Image2-Skill

All skills in this repo
  1. gpt-imagewuyoscar/GPT-Image2-Skill

    Generate or edit images with GPT Image 2 or 2.5 through the packaged CLI and Reference Gallery. Use for image requests including imprecise 'GPT 2.5' model names, posters, typography, reference edits, and inpainting; resolve the model choice before generation.

    5.8k162 tok idle · 2,380 in use