图片与海报
Social Image Ad
A scroll-stopping static ad for Instagram, Facebook or Pinterest with a clear focal point and space for a headline.
OraclipSkill
图片与海报可直接使用
Social Image Ad
A scroll-stopping static ad for Instagram, Facebook or Pinterest with a clear focal point and space for a headline.
适合场景
SocialImage
SKILL.md
--- name: 'social-image-ad' description: 'A scroll-stopping static ad for Instagram, Facebook or Pinterest with a clear focal point and space for a headline.' --- # Social Image Ad Designs a single feed ad image with a bold visual idea and clean negative space for copy. ## Use it for - Meta, Instagram, Pinterest and display ads - Promotions, launches and seasonal campaigns ## Inputs - Product or offer - Optional: headline text (max ~6 words), brand colors, audience ## Defaults - Call `generate_image`. - Aspect ratio 4:5 (feed); offer 9:16 for stories and 1:1. - Only render text when the user gives it; otherwise leave empty space for copy and say so. ## Prompt recipe `Static social media ad, [one bold visual idea featuring product], [brand colors] palette, strong single focal point, generous empty space at [top/left] for a headline, high contrast, clean modern composition, [style: photographic / 3D render / flat graphic]` + optionally `with the headline "[text]" in bold sans-serif at the top` ## Example prompt ```text Static social media ad, a pair of white running shoes bursting out of a splash of bright orange paint against a deep navy background, strong single focal point, generous empty space at the top for a headline, high contrast, clean modern composition, crisp photographic render, with the headline "Run Brighter" in bold white sans-serif at the top ``` ## Variations to offer - Three different visual hooks for A/B testing - Same ad in 9:16 and 1:1 - A minimal text-free version ## Avoid - Long sentences in the image - Fake discounts or prices the user did not give - Tiny details that disappear on a phone screen ## Working rules - Reply in the user's language; write the generation prompt itself in English. - Ask at most one short question, and only when a missing detail would change the result (for example the product is unknown and no image is attached). Otherwise use the defaults in this skill and go. - Keep one generation call per requested output. A video clip is one continuous shot; for multi-shot ideas, render the most important shot first and offer the next ones. - Respect the composer settings (model, duration, resolution) unless this skill says otherwise or the user asks for something different. - Never imitate a real person's likeness, a living artist's signature style, or a trademarked character or logo the user does not own. Swap them for an original description. - After the result, offer two concrete follow-ups (for example a different angle, aspect ratio, or mood).