Menu
Generate, edit, or refine images using Gemini or OpenAI. Auto-detects intent from context.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
Run quality checks on a task's implementation before it closes to done. Triggers: 'is this ready?', 'run the review gate', 'check my work', 'I think this is done'. Reviews code + spec adherence, runs tests/build. Design review uses taskmaster:spec-review.
Log a lightweight idea into .taskmaster/ideas/. Triggers: '/add-idea', 'save this as an idea', 'remember this idea', 'log this idea', or something worth keeping, not yet a task. Only way - not backlog_idea_create directly.
Log/update/close project bugs in .taskmaster/bugs/. Triggers: 'log a bug', 'this is a bug', 'track this defect', 'I found a bug', 'shelve this for later', 'list open bugs', 'promote B-XX', 'close B-XX'. Also for ambiguous 'issue' phrasing lacking recurring/systemic/outstanding evidence - falls back here. Only correct way to transition a project Bug.
Scan the codebase for TODO/FIXME/HACK/XXX and cross-reference the backlog. Triggers: 'check todos', 'are my todos tracked', 'scan for todos', 'todo audit', 'what's untracked'.
Write/resolve/drop project decisions in .taskmaster/decisions/. Invoke when about to write an inline option menu - route here, not Options: in chat. Also 'choose between', 'pick an option', 'decide on', 'open question', 'resolve DEC-X', 'drop DEC-X'. Only correct way - do not call backlog_decision_create directly.
Close out a work session. Triggers: 'end session', 'I'm done for today', 'let's wrap up', 'mark this task done', 'save progress'. ONLY correct way to mark tasks done/in-review with a session record.
| name | image-gen |
| description | Generate, edit, or refine images using Gemini or OpenAI. Auto-detects intent from context. |
You have access to image generation via a Node.js script. Supports two backends: Gemini (default) and OpenAI (GPT Image 2, with GPT Image 1.5 retained for transparency).
Analyze the user's request and route:
| Signal | Action |
|---|---|
| No existing image referenced | Generate — invoke image-gen:generate |
| User references an existing image file path | Edit — invoke image-gen:edit |
| User says "refine", "iterate", "adjust", or references a previously generated image from this session | Refine — invoke image-gen:refine |
| Need | Backend | Why |
|---|---|---|
| General image generation | gemini (default) | Fast, free tier, good quality |
| Transparent backgrounds (game assets, sprites, icons) | openai with --transparent | Auto-uses gpt-image-1.5 (native alpha). gpt-image-2 does not support transparency — script falls back automatically. |
| Transparent + Gemini style | gemini with --transparent | Two-pass alpha extraction (slower, 2 API calls) |
| Highest quality generation | openai with --quality high | gpt-image-2 is the new state-of-the-art default |
GEMINI_API_KEY environment variableOPENAI_API_KEY environment variablenpm install --prefix "${SKILL_DIR}/../../src" if node_modules doesn't existScript location: ${SKILL_DIR}/../../src/generate.mjs
Full usage:
node "${SKILL_DIR}/../../src/generate.mjs" \
--prompt "..." \
[--output path] \
[--input path] \
[--backend gemini|openai] \
[--model name] \
[--aspect ratio] \
[--size size] \
[--transparent] \
[--quality low|medium|high]
Defaults: backend=gemini, model=auto (gemini-3.1-flash-image-preview or gpt-image-2), aspect=1:1, size=1K, quality=medium
Gemini models:
gemini-3.1-flash-image-preview — Fast, good for iteration (default)gemini-3-pro-image-preview — Highest quality, use for final/polished assetsgemini-2.5-flash-image — High-volume, low-latencyOpenAI models:
gpt-image-2 — State-of-the-art, fastest, supports flexible sizes up to 3840px and quality=auto (default).gpt-image-1.5 — Retained for transparent backgrounds. gpt-image-2 does not support background=transparent, so the script auto-falls-back to gpt-image-1.5 whenever --transparent is set. You can also pin it explicitly with --model gpt-image-1.5.Aspect ratios: 1:1, 2:3, 3:2, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9
Gemini sizes: 512px, 1K, 2K, 4K OpenAI sizes: Mapped from aspect ratio (1024x1024, 1536x1024, 1024x1536)
OpenAI quality (1024x1024):
gpt-image-2: low ~$0.005, medium ~$0.041, high ~$0.211, auto (default for the model)gpt-image-1.5: low $0.009, medium $0.034, high $0.133After routing, invoke the appropriate sub-skill. Do NOT generate images directly from this router — always delegate to a sub-skill.