GPT Image 2 Skill
162-prompt gallery plus a real CLI for GPT Image 2 generation and editing
Test report
- Verdict
- Tested · Works
- Score
- Tested
- Jul 16, 2026
- Environment
- Claude Code 2.x (agent harness)
- Upstream re-checked
- Aug 10, 2026 · 0db0ea3
Task: create a prompt for a vertical poster for an autumn milk tea launch with an exact Chinese copyright string, prices, and promotion dates, and run it through the standard CLI. Baseline prompt without skill — 45 words and 0 quoted strings, meaning for a poster whose entire purpose is precise Chinese text, not a single displayed word was fixed. Skill prompt (gallery-typography-and-posters.md No. 33 + craft.md) — 198 words, 19 exact quoted strings, 3:4 canvas declared first, clear hierarchy, small font module, 'Simplified Chinese' and an avoid-string against garbled characters. Plumbing checked live: uv run scripts/generate.py from the installed folder pulled the CLI via uvx and exited with code 2 and precisely the OPENAI_API_KEY message described in SKILL.md; '-m without -i' gave '--mask requires --image', a non-existent -i gave '--image not found', and with a dummy key, it reached a real 401 from OpenAI — meaning the SDK path and endpoint routing are working. Claim for 162 prompts verified by count: exactly 162 numbered entries, all 31 category files in place. HONEST LIMITATION: no actual image was generated — no valid OPENAI_API_KEY, so the final pixel result was not verified, hence 8/10, not higher.
Scored on four weighted criteria — install, triggering, output vs. baseline, docs. How scoring works
- Installs cleanly 5/5
- Triggers reliably 5/5
- Output vs. baseline 8/10
- Docs & honesty 4/5
What GPT Image 2 Skill does
Routes image requests through a 162-prompt reference gallery and a prompt-craft checklist, then calls a packaged CLI for GPT Image 2 generation, reference editing and inpainting. Triggers when you ask to generate, render or edit images, posters, typography, Chinese text, UI mockups or diagrams.
How to install GPT Image 2 Skill
git clone https://github.com/wuyoscar/GPT-Image2-Skill.git
mkdir -p ~/.claude/skills
cp -r GPT-Image2-Skill/skills/gpt-image ~/.claude/skills/gpt-image
Skills live in ~/.claude/skills/ (global) or .claude/skills/
(per-project). Restart Claude Code after installing.
Commands — how to trigger GPT Image 2 Skill
-
/gpt-image162-prompt gallery plus a real CLI for GPT Image 2 generation and editing
It also activates on plain-language prompts like these:
-
Generate a poster for our milk tea launch with exact Chinese pricing text -
Create a UI mockup image with GPT Image 2 for this app screen -
Edit this reference photo by inpainting a new background with GPT Image 2
Frequently asked questions
- Is the GPT Image 2 Skill skill free?
- Yes. The skill itself is free from wuyoscar/GPT-Image2-Skill. SkillProof publishes the install command and an independent test verdict at no cost.
- Does GPT Image 2 Skill work with Claude Code?
- We tested it with Claude Code 2.x (agent harness) on Jul 16, 2026. Verdict: Tested · Works. Task: create a prompt for a vertical poster for an autumn milk tea launch with an exact Chinese copyright string, prices, and promotion dates, and run it through the standard CLI. Baseline prompt without skill — 45 words and 0 quoted strings, meaning for a poster whose entire purpose is precise Chinese text, not a single displayed word was fixed. Skill prompt (gallery-typography-and-posters.md No. 33 + craft.md) — 198 words, 19 exact quoted strings, 3:4 canvas declared first, clear hierarchy, small font module, 'Simplified Chinese' and an avoid-string against garbled characters. Plumbing checked live: uv run scripts/generate.py from the installed folder pulled the CLI via uvx and exited with code 2 and precisely the OPENAI_API_KEY message described in SKILL.md; '-m without -i' gave '--mask requires --image', a non-existent -i gave '--image not found', and with a dummy key, it reached a real 401 from OpenAI — meaning the SDK path and endpoint routing are working. Claim for 162 prompts verified by count: exactly 162 numbered entries, all 31 category files in place. HONEST LIMITATION: no actual image was generated — no valid OPENAI_API_KEY, so the final pixel result was not verified, hence 8/10, not higher.
- What is the GPT Image 2 Skill SkillProof Score?
- 8.8/10 — installs cleanly 5/5, triggers reliably 5/5, output vs. baseline 8/10, docs & honesty 4/5.
- How do I install GPT Image 2 Skill?
- Copy the install command from this page, run it in your terminal, and restart Claude Code. Skills live in ~/.claude/skills/ (global) or .claude/skills/ inside a project.
- Can I use GPT Image 2 Skill with Cursor, Copilot, Gemini CLI, Codex or other AI tools?
- The SKILL.md format is native to Claude (Claude Code, Desktop, claude.ai). The instructions inside adapt to other assistants: Cursor rules, GitHub Copilot instructions, Windsurf rules, Custom GPTs, AGENTS.md for OpenAI Codex, and GEMINI.md for Google Gemini CLI — our conversion guides cover each, and the free converter on the tools page does the wrapping for you.