ck:ai-artist

Generate stylized and photorealistic images from curated Nano Banana prompts.

1|Updated Mar 6, 2026
One-click install
npx skills add https://github.com/quanganh208/cookmate --skill ck-ai-artist-quanganh208
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: ck:ai-artist
Source: https://github.com/quanganh208/cookmate/tree/main/.opencode/skills/ai-artist
Command: npx skills add https://github.com/quanganh208/cookmate --skill ck-ai-artist-quanganh208

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires google-genai, python-dotenv, and includes scripts (resource) and references (resource) components.

What problem does it solve?

This skill reduces the time and guesswork required to craft high-quality image prompts and iterate visual concepts by providing a searchable, curated prompt database and automated generation workflows tied to Nano Banana (Gemini) models.

Core Features & Use Cases

  • Prompt search & ranking: BM25-powered search over 129 curated prompts to find best matches for a concept.
  • Three creative modes: search (best match), creative (remix top matches), and wild (random artistic transformation), plus an all mode to generate variations.
  • Validation workflow: mandatory interview to capture style, mood, colors and aspect ratio with an option to skip for defaults.
  • Generation orchestration: prompt adaptation, model and aspect-ratio selection, dry-run and verbose inspection, and direct generation via Gemini (google-genai).
  • Developer tools: scripts for searching, extracting prompts, and generating images; references for advanced prompting and technique tuning.

Quick Start

Generate a cinematic tech conference banner in search mode and save the result as banner.png.

Frequently Asked Questions about ck:ai-artist

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

FAQPage Schema
How do I generate high-quality images from text prompts without manual trial and error?

Generate high-quality images by searching a curated database of 129 Nano Banana prompts using BM25 ranking to find the best matches for your concept, then use Gemini models to produce polished outputs for avatars, banners, and product showcases.

Can I use Gemini models to create stylized art and photorealistic outputs for social media assets?

Yes, you can use Gemini models via the google-genai API to create stylized art and photorealistic outputs for social media assets, applying prompt adaptation and aspect ratio selection to refine the final generated images.

How do I search and remix existing image prompts to iterate visual concepts faster?

Search and remix existing image prompts by using the creative mode to remix top BM25 search matches, or use wild mode for random artistic transformations, reducing the guesswork required to iterate visual concepts.

Do I need a GEMINI_API_KEY to run image generation workflows, and can I test without making API calls?

Yes, you need a GEMINI_API_KEY for direct image generation via google-genai, but you can use the dry-run option and validation workflow to inspect prompt adaptations and test the generation orchestration without making actual API calls.

What is the best way to select aspect ratios and adapt prompts for different marketing use cases?

The best way to select aspect ratios and adapt prompts is through the mandatory validation interview, which captures your desired style, mood, colors, and aspect ratio, or you can skip it to use default settings for marketing assets.

Why does my prompt search return irrelevant matches when generating images from text descriptions?

Prompt search returns irrelevant matches when keywords are too broad, but BM25-powered search over the curated database ranks 129 specific prompts by relevance to your concept, improving match accuracy for the generation workflow.