Prompt library
PROMPTS
1328 hand-tested AI prompts across image, video, and music — each with full context, variables, and a sample output. Content is authored in Traditional Chinese; the UI here is English.
02Tool
03Use case
IMAGE
免費01IMAGE／Midjourney v7
2026-06-15
Authentic Taiwan Street Scenes: Night Markets, Not Generic Asia
Midjourney v7 uses a precise dictionary of Taiwan-specific elements (vertical neon signage, red lanterns, corrugated-metal arcades, scooter swarms) to avoid generating scenes that read as Japan or Hong Kong instead — for local-revitalization and food/tourism material.
IntermediateGen in-site
Read
免費02IMAGE／Midjourney v7
2026-04-10
Midjourney Cinematic Portrait: Soft Rim-Lit Subject
A magazine-cover-like portrait style, suited to personal-brand avatars and portfolio hero images.
SAMPLE雜誌封面級人像：側臉 3/4 角度，柔光從左前方打下，右臉輕陰影帶層次，眼神看向鏡頭外，深色羊毛大衣配米白襯衫，背景是低飽和的莫蘭迪藍灰色調微微失焦，髮絲邊緣帶逆光金邊，皮膚紋理保留但無瑕疵，可直接用於個人品牌 LinkedIn 或作品集。
BeginnerGen in-site
Read