From google-image-gen
Generate images using Google's Gemini API. Use this skill when the user wants to generate, create, or edit images with AI. Keywords: image generation, create image, generate picture, AI art, edit image, Gemini, icon, render, illustration.
npx claudepluginhub ypfaff/google-image-gen-plugin --plugin google-image-genThis skill is limited to using the following tools:
Generate images from text prompts using Google's Gemini API.
Guides Next.js Cache Components and Partial Prerendering (PPR) with cacheComponents enabled. Implements 'use cache', cacheLife(), cacheTag(), revalidateTag(), static/dynamic optimization, and cache debugging.
Guides building MCP servers enabling LLMs to interact with external services via tools. Covers best practices, TypeScript/Node (MCP SDK), Python (FastMCP).
Generates original PNG/PDF visual art via design philosophy manifestos for posters, graphics, and static designs on user request.
Generate images from text prompts using Google's Gemini API.
Run these commands once at the start of a session:
${CLAUDE_PLUGIN_ROOT}/scripts/check_env.sh
cd ${CLAUDE_PLUGIN_ROOT} && uv sync && cd -
If the environment check fails, the user needs to configure their API key.
Recommended: Create ~/.config/google-image-gen/.env with:
GOOGLE_AI_API_KEY=your_key_here
Get an API key from https://aistudio.google.com/apikey
Alternative: Export as environment variable: export GOOGLE_AI_API_KEY=your_key
Important: The plugin runs from its own directory. Use --cwd to ensure output files land in the user's project,
not the plugin cache.
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" <output_path> "<prompt>" [options]
Note: When the user specifies a path like output.png or images/photo.png, pass it as-is — the --cwd parameter
ensures it resolves relative to the user's project root.
| Option | Short | Description |
|---|---|---|
--style | -s | Style template (.md file with {subject} placeholder) |
--ref | -r | Reference image for style (repeatable, max 14) |
--edit | -e | Edit existing image instead of generating |
--aspect | -a | Aspect ratio: 1:1, 3:4, 4:3, 4:5, 5:4, 9:16, 16:9, 21:9 |
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" output.png "A red apple on a wooden table"
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" images/output.png "A red apple"
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" thumb.png "Mountain landscape" --aspect 16:9
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" output.png "Change the sky to sunset" --edit input.png
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" output.png "Same style but with a car" --ref reference.png
Generates numbered outputs (output_1.png, output_2.png, etc.):
ORIG_CWD="$(pwd)" && cd ${CLAUDE_PLUGIN_ROOT} && uv run python main.py --cwd "$ORIG_CWD" output.png "cat" "dog" "bird"