AI Image Prompts skill
A skill that searches a library of 10,000+ image prompts, shows sample images and tailors the chosen prompt to your article or script
- 1,350 ★ on the repository
- Last commit
- MIT
Minimal access
Why this level
- On install the postinstall script downloads library files from GitHub
- Sample images in replies open from external links
Install
- Works in
Manual install
npx skills i YouMind-OpenLab/ai-image-prompts-skillThe universal installer detects the agent itself. The prompt library downloads at postinstall.
Access and checks
How we reviewMinimal accessThe tool runs nothing and makes no network calls.
- On install the postinstall script downloads library files from GitHub
- Sample images in replies open from external links
This is third-party code. Review the repository files before installing.
What it does
AI Image Prompts turns a description of the picture you need into a short list of ready prompts. The agent searches JSON files in the references folder, sorted into 11 categories: social posts, product marketing, avatars, posters, infographics, game assets, comics, YouTube thumbnails, app and web design. The answer holds up to three options, each with a translated title, a short prompt excerpt, a link to the full text and a sample image. In the second mode you give an article or script: the skill reads its theme and tone, offers matching styles and, after you pick one, builds a prompt around your material. The prompts are not tied to one model. The setup.js script downloads the library on install, and the repository refreshes it twice a day.
Who it is for. For writers, designers and marketers who need covers, thumbnails and illustrations and do not want to write prompts from scratch.
Good fit when
- You need a cover for an article, for example about remote work, and want to pick a style from samples
- You need a thumbnail for a YouTube video, such as a tech review
- You need a product photo on a white background for a store listing
- You need an avatar or portrait in a given style, say cyberpunk
- You try different image generators and want prompts from one library
Not a fit when
- You need the image itself: the skill picks and tailors prompts, and your model draws the picture
- The install runs without access to GitHub: the postinstall script downloads the library
- You need prompts for code or text: the library holds image prompts only
Example request
Find a prompt for a cover of an article on remote work productivity in a minimalist style and show three options with samplesLimitations
Install needs access to GitHub: postinstall pulls the library, and you refresh it by hand with node scripts/setup.js --force or pnpm run sync. By the skill's own rules, prompts without a sample image are skipped. Replies come in the user's language, but the prompt for generation stays in English. The prompts are collected from community posts on YouMind, and the repository is MIT licensed. The prompt count in the repository changes: the library grows along with community posts.
How to disable. Delete the ai-image-prompts-skill folder from the agent's skills directory. The downloaded files live in references inside the same folder and go with it.
FAQ
How do I install AI Image Prompts in Claude Code?
Run npx skills i YouMind-OpenLab/ai-image-prompts-skill. The same command works for Cursor, Codex, Gemini CLI and Windsurf, and OpenClaw has clawhub install ai-image-prompts. Postinstall downloads the library.
Which models do these prompts work with?
The README names Nano Banana Pro and Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E 3, Flux and Stable Diffusion. The prompts are written in English and do not depend on a model.
How is AI Image Prompts different from AI Image Prompt Recommender?
Both skills come from YouMind. AI Image Prompt Recommender targets Nano Banana Pro, while AI Image Prompts builds a library of more than 10,000 prompts for any model and splits it into 11 categories.
How do I refresh the prompt library?
Run pnpm run sync or node scripts/setup.js --force. node scripts/setup.js --check quietly tells you whether the data is stale. In the repository the library refreshes twice a day through GitHub Actions.
What does the content illustration mode do?
You paste an article, a video script or podcast notes. The skill reads the theme, tone and audience, searches matching templates, lets you pick a style and, once you do, builds a prompt around your material after a few clarifying questions.
Discussion
No questions or reviews yet.
Related
SkillsNano Banana Pro Prompts Recommend
AI Image Prompt Recommender
A skill that picks ready image-generation prompts from a large library, shows sample images and assembles a prompt for your text
A set of skills for Claude Code and Codex that assemble detailed image prompts from an element library and composition schemes
A skill that writes prompts for video, image and music generation on 14 platforms and can submit them in the browser
A library of 150 GPT Image prompts, each credited to its author with a link to the post, grouped into nine use cases
GPT-Image2 Style Library
gpt-image-2-style-library
A skill picks a matching template and style from a library of GPT-Image2 cases and assembles it into a finished prompt
A Claude Code skill that turns an idea into a five-stage prompt for Sora, Kling, Veo and Seedance, with templates for 21 genres
Comparison with similar tools
Entries that share the main search query, and how they differ.
| Entry | Official | Repository stars | Main difference |
|---|---|---|---|
| AI Image Prompts (this entry) | no | 1,350 | |
| Nano Banana Pro Prompts Recommend | no | 1,892 | an earlier YouMind skill aimed at Nano Banana Pro, while this one works with any model |
| GPT Image 2 Prompts by YouMind | no | 10,192 | a prompt library for GPT Image 2 as a list, with no agent skill |
README in short
The README explains that the skill gives an agent search over a curated library of more than 10,000 prompts and lists the models they suit. It then covers install for OpenClaw, Claude Code and other agents, two modes (direct search and illustration from your content), a table of 11 categories with prompt counts, the data source on YouMind and how to refresh the library by hand.
README badge
Original SKILL.md
--- name: ai-image-prompts-skill description: | Recommend curated prompts from a 10,000+ real-world image generation prompt library. Works with ANY AI image model — Nano Banana Pro, Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E 3, Flux, Stable Diffusion, and more. Use this skill when users want to: - Find proven image generation prompts (any model) - Get prompt inspiration for portraits, products, social media, posters, etc. - Create illustrations for articles, videos, podcasts, or marketing content - Browse categorized prompt templates with sample images - Translate and understand prompt techniques platforms: - openclaw - claude-code - cursor - co