← ClaudeAtlas

varg-ad-variantslisted

Turns one winning video ad into 20 ad variants that Meta's Andromeda and TikTok treat as different creatives: tag the winner with varg video_analyze, plan two layers (about 7 new concepts that each change message, visual execution or format, plus about 13 single-variable iterations, mostly new hooks), generate the new openers and concepts on varg (Seedance 2.5, Kling 4, Nano Banana, ElevenLabs), swap hooks onto the winner with a tested FFmpeg script (word-boundary cut, normalised parts, 20 ms fade, manifest), re-caption every variant, and export names and a test plan. Use when a seller asks for ad variations, more creatives, hook variants, "scale my winning ad", "make 20 versions of this ad", creative testing volume for Meta or TikTok, or new hooks for an ad that is fatiguing. Not for the testing strategy alone (ad-creative-testing), finding winners from competitors (varg-ad-library-remix), or a brand-new single ad.
vargHQ/skills · ★ 38 · Code & Development · score 86
Install: claude install-skill vargHQ/skills
# One winning ad into 20 variants Since Meta's Andromeda update, ads that look alike are grouped and compete as one ("Entity ID" in agency talk): a new caption, colour or music track is not a new creative. Sellers now test 20–40 creatives a week at about $20/day each and refresh every 2–3 weeks ([theoptimizer](https://theoptimizer.io/blog/how-to-test-ad-creatives-on-meta-after-the-andromeda-update-2026-playbook), [segwise](https://segwise.ai/blog/meta-andromeda-update-creative-strategy-2026)). The "suppressed above 60% similarity" thresholds are repeated by agencies but not published by Meta. ## 1. Tag the winner `video_analyze` on the winner (about 5 credits; prompt in `varg-ad-library-remix`) → hook, angle, format, emotional driver, persona, proof, CTA. You scale the mechanism behind the ad, not the file. ## 2. Plan two layers (20 in total) **Layer A, ~7 new concepts.** Each must pass the three-question check (two "no" answers = the same Entity): *Is the message different? Is the visual execution different, not just the text? Is the format different?* Moves: same script with a new persona (female → male → couple → parent), same structure with a new script, same concept in a new format (UGC → unboxing → demo → mascot song → problem/solution), same script with new b-roll. **Layer B, ~13 iterations, one variable each.** Mostly new hooks (3–5 per concept, ≤3 s), then new CTAs or offers. If two fields change between variants, the plan is wrong ([Pixis](https://pixis.ai/blo