A Python-like prompt structure designed to generate high-CTR YouTube thumbnails. It includes components for hero treatment, supporting elements, text blocks, and curiosity gaps, along with verification assertions for mobile legibility, focal count, text restraint, authenticity, and asymmetry. The prompt is flexible and can work with or without additional style pools.
class ThumbnailEngine: def __init__(self, SUBJECT, variables=None): self.hook = diagnose_hook(SUBJECT) self.picks = sample_pools(self.hook) # 2 from relevant pools def compose(self): return Thumbnail( hero = self.picks.hero_treatment, support = self.picks.reaction_or_proof, text = TextBlock(words=self.picks.text_hook, max_words=6), gap = self.picks.curiosity_gap ) def render(self): return high_ctr_image( subject=self.compose(), environment=self.picks.context_world, lighting=self.picks.color_contrast, camera=self.picks.perspective ) def verify(self, img): assert mobile_legible(img) and focal_count(img) <= 3, "T₂: Mobile Hierarchy" assert curiosity_gap_present(img) and not fully_explained(img), "T₁: Curiosity Gap" assert text_word_count(img) <= 6 and not title_repeated(img), "T₅: Text Restraint" assert authentic_environment(img) or justified_cgi(img), "T₃: Authenticity" assert asymmetric_tension(img) if is_comparison(img) else True, "T₄: Asymmetry"
Click the copy button above to copy the prompt to your clipboard.
Open your preferred image generation tool like Midjourney, DALL-E, or Flux.
Paste the prompt into your tool, adjust settings if needed, and generate.
Try our AI Studio with this prompt or create something entirely new.