10,000+ prompt library works across every major image generator

YouMind-OpenLab put the whole thing on GitHub under MIT, which means anyone can pull the repository, fork it, or fold it into a paid product without asking permission first. The count sits above 10,000 prompts now, split across eleven use-case categories rather than dumped into one giant list. Social media posts make up the biggest chunk by far at 6,382, followed by product marketing at 3,709. After that it thins out fast: profile and avatar shots, posters and flyers, infographics, e-commerce product images, game assets, comic panels, YouTube thumbnails, app and web design work, plus a few hundred that haven’t been sorted into a category yet.

None of it is locked to one image generator. The prompts are written to work across Nano Banana Pro, Nano Banana 2, Seedream 5.0, GPT Image 1.5, Midjourney, DALL-E 3, Flux, and Stable Diffusion, so switching tools doesn’t mean rewriting a prompt from scratch. Every entry ships with a translated description, sample preview text, an actual generated image to show what the prompt produces, and a set of customization options rather than a flat string of text with no context attached.

The content itself is community-sourced and pulled together by YouMind.com, then synced into the repo automatically twice a day through GitHub Actions — so this isn’t a static dump that stops growing the day someone gets bored of maintaining it. That update cadence also means the category breakdown above is a snapshot, not a fixed ceiling; whichever categories the community leans into next will keep shifting those numbers upward. On the technical side there’s a proper build setup behind it too: TypeScript configuration, a pnpm-managed package.json, and sync scripts running the whole pipeline.

Beyond just being a prompt library to browse, it’s built to work as a skill inside AI assistants including Claude, OpenClaw, and Cursor, with semantic search built in and a remix function that can spin up a new custom prompt out of whatever material a user feeds it. The whole thing sits in the ai-image-prompts-skill repository under YouMind-OpenLab on GitHub, free to clone right now.