skills.rest
joeseesun
joeseesunCOMMUNITY
Explore all skills by this author →
𝕏@vista8GitHub

image-gen

Generate AI images from text prompts via the ListenHub CLI with configurable model, resolution, and aspect ratio.

335|37|Updated Jul 18, 2026
Search Tags:#image generation#text-to-image#ai art#prompt#reference images#aspect ratio#listenhub
One-click install
npx skills add https://github.com/joeseesun/qiaomu-cut-skill --skill image-gen-joeseesun
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: image-gen
Source: https://github.com/joeseesun/qiaomu-cut-skill/tree/main/vendor/marswaveai-skills/image-gen
Command: npx skills add https://github.com/joeseesun/qiaomu-cut-skill --skill image-gen-joeseesun

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve? Creating custom images for covers, illustrations, or concept art normally requires switching to a separate design tool or image service. This Skill turns a plain text description into a generated image directly inside the agent workflow, with guided parameter selection and local file output. ## Core Features & Use Cases - Text-to-Image Generation: Submit prompts to gemini-3-pro-image or gemini-3.1-flash-image models with 1K/2K/4K resolution and multiple aspect ratios including extreme ratios like 1:4 and 8:1 on flash. - Reference Image Guidance: Pass up to 5 local files or URLs as style references via --reference flags. - Guided Confirmation Flow: Collects prompt, model, resolution, and ratio step by step, then requires explicit user confirmation before generation. - Use Case: A user asks for a cyberpunk city cover image; the Skill confirms 2K resolution and 16:9 ratio, runs the ListenHub CLI, and saves the result to .listenhub/image-gen/ with inline preview. ## Quick Start Ask the agent to generate an image of your subject, then answer the prompts for model, resolution, and aspect ratio to receive the saved image file.

Frequently Asked Questions about image-gen

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I generate an AI image from a text prompt?▼

Describe the image you want, then choose a model, resolution, and aspect ratio when prompted. The Skill builds a listenhub image create command with your parameters and runs it only after you explicitly confirm the summarized choices.

What is the difference between the pro and flash image models?▼

The pro model (gemini-3-pro-image) offers higher quality, while flash (gemini-3.1-flash-image) is faster and cheaper. Flash also unlocks extreme aspect ratios such as 1:4, 4:1, 1:8, and 8:1 that pro does not support.

Can I use reference images to guide AI image generation?▼

Yes, you can provide up to 5 reference images as local file paths or URLs in jpg, png, webp, or gif format, each under 10MB. Each reference is passed to the CLI with a --reference flag to guide the visual style.

Does image generation work without ListenHub CLI authentication?▼

No, the Skill checks CLI authentication first and auto-installs or auto-logs in if needed. It supports two auth modes: OpenAPI key via LISTENHUB_API_KEY or OAuth login, which determine whether commands use listenhub openapi image or listenhub image.

Where are generated images saved?▼

Images are saved to .listenhub/image-gen/YYYY-MM-DD-{jobId}/ directories, never to ~/Downloads. Depending on the configured outputMode, the image can also be displayed inline in the conversation via a temporary download.

What are the limitations of this image generation approach?▼

It cannot edit existing images, generate audio or video, or extract content from URLs. Prompts should be written in English since the underlying models are trained on English descriptions, and generation requires explicit user confirmation before running.

Related Skills

Explore More Design & Creative Skills→
MattSenterMattSenter

senternet-site-gcloud-auth

Installs and authenticates the Google Cloud SDK and Firebase CLI for site setup workflows.

Community
Intermediate
hoanghd218hoanghd218
3

mkt-broll-image

Generate vertical AI b-roll images for short-form video production workflows.

Community
Intermediate
almazomalmazom

minimax-multimodal-toolkit

Orchestrate multimedia generation and processing via MiniMax multimodal APIs.

Community
Advanced
hsliuustc0106hsliuustc0106
84

vllm-omni-image-gen

Generate and edit images from text prompts using vLLM-Omni diffusion models.

Community
Advanced