Turn 1 idea into 7 assets without rewriting from scratch
Converts a single "pillar" idea into platform-native variations (short video script, carousel outline, X thread, LinkedIn post, YouTube description, Pinterest pin copy). Includes a per-platform packaging guide so each version matches the platform's hook style, pacing, CTA norms, and length expectations. Designed for creators who find repurposing doubles their workload instead of reducing it.
Create a skill called "Repurpose Without Double Work". Input can be: transcript, outline, thread, or a single core idea. Ask for: - target platforms (choose any) - my tone and primary CTA - how many assets I want (3–10) - audience intent per platform if it differs (learn, laugh, buy, compare, follow) Output: - For each platform: (1) platform-native hook, (2) structure/outline, (3) caption/copy, (4) CTA, (5) packaging note (recommended length, pacing, keywords/hashtags). - A "Keep / Change" cheat sheet summarizing format rules per platform in one table. - Risk notes: what hurts performance on each platform (e.g., external links on TikTok). Guardrail: Do not copy/paste across platforms; each version must be adapted to that platform's behavior and audience expectations.
You provide one core idea (or one piece of content). The skill produces a repurposing pack
with platform-specific edits: different hooks, length, structure, and CTA.
Output: "IG carousel outline + TikTok script + LinkedIn post + X thread, each with platform-native hooks and CTA."
Output: "LinkedIn rewrite + IG caption + short-form script with 3 hooks."
Output: "TikTok fast-cut script, IG carousel, LinkedIn story version, X thread — each with platform-specific packaging notes."
One idea, native variants for every platform
Algorithms reward platform-native content and can throttle copy-paste distribution. This recipe takes one "core insight" and produces native variants for each platform (hooks, length, formatting, CTA), plus a publishing and engagement plan.
Wikipedia-grade AI pattern removal
Comprehensive AI writing cleanup based on Wikipedia's WikiProject AI Cleanup guidelines. Catches 24+ distinct patterns including inflated symbolism, em dash overuse, rule of three, copula avoidance, and sycophantic tone.
Unblock pip installs that trigger native compilation on Windows.
Many scientific Python packages still require native compilation on Windows when wheels are unavailable; this recipe turns the "install build tools" advice into a reproducible checklist.
Give meaningful feedback without writing essays on every paper
Detailed feedback matters, but writing individualized paragraphs on 150 papers is not sustainable. This recipe builds a coded feedback system that's fast for you and clear for students — so grading doesn't consume every evening.