Content Repurpose Engine

content-repurpose-engine

Purpose — Take one insight / learning / event and produce platform-native versions for X (EN), X (ZH), long-form blog, LinkedIn, and video hook — each adapted to the platform's format, not just reposted.

Triggers — "repurpose this into threads", "turn this into content", "multi-platform post this insight"

Why this matters — Reposting the same 280 chars to 5 platforms dilutes reach on all 5. Platform-native adaptation takes 3× the effort but 5–10× the traction.

---

Inputs

• Source insight — 1 paragraph + (optionally) supporting data / screenshot / code snippet
• Target platforms — subset of: X-EN, X-ZH, LinkedIn, blog, video hook (script only)
• Voice sample (optional but strongly recommended) — 2–3 past posts in your voice for tone calibration

Outputs

Per target platform, one draft file:

• x-en.md — single-tweet + thread variant
• x-zh.md — single-tweet + thread variant
• linkedin.md — 1 long-form post (800–1200 chars)
• blog.md — 600–1200 words, SEO-conscious heading structure
• video-hook.md — first 8 seconds script + outline for 60-second video

Plus cross-posting-notes.md — what to post where, in what order, with what interval.

Steps

1. Read source + voice samples. Identify the ONE point worth making. Kill any urge to cram multiple points.
2. Draft each platform sequentially, not in parallel. Each platform has its own shape:

X-EN (single tweet)

• Max 280 chars
• Hook in first line. No "As AI evolves…" openings.
• Single concrete claim + evidence, OR one counter-intuitive take
• End with the most quotable sentence (not a question)

X-EN (thread)

• 5–9 tweets
• Tweet 1 = strong hook + promise payoff
• Tweets 2–N = one claim + one line of evidence per tweet
• Last tweet = either a single practical takeaway or a soft CTA (link / follow)

X-ZH

• 不是 X-EN 的翻译。重写
• 250–280 字（中文 tweet 字数不同）
• 保留技术名词的英文原文，行文口语化
• 避免"大家好"、"今天给大家分享"等套话

LinkedIn

• 800–1200 chars
• Start with a strong 1-line hook (first 210 chars shown before "see more")
• Use short paragraphs (1–2 lines each)
• 1 concrete story > 3 abstract claims
• Soft CTA at end (not "DM me for consulting")

Blog

• 600–1200 words
• H2 structure: Problem → Framework → Example → Takeaway
• First paragraph = TL;DR in 2 sentences
• Canonical link to source insight if it's from a repo / case study

Video hook

• First 8 seconds must state the counter-intuitive claim
• Outline: hook (8s) → setup (20s) → demonstration (25s) → takeaway (7s)
• Include B-roll / screen-capture notes

Quality bar

• [ ] No platform post reuses another platform's opening line verbatim
• [ ] X-ZH is not a machine translation of X-EN
• [ ] LinkedIn doesn't open with "🚀 Excited to share…"
• [ ] Blog has canonical URL / citations for any non-original data
• [ ] Cross-posting notes specify at least 6-hour gaps between platforms

Anti-patterns

• Don't post to all platforms in the same hour. Spread by ≥6h to let each breathe.
• Don't reply to your own thread on day 2 with "update:". Post a new thread instead.
• Don't use emojis as bullet points on LinkedIn. It reads as LinkedIn-influencer slop.
• Don't write X-ZH as a translation. Native rewriting is 10× the impact.

Related

• Skill: [growth-experiment-template](../growth-experiment-template) — for measuring which format worked
• Playbook: (none yet — TBD)

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

• Author: Celina-create
• Source: Celina-create/X-Studio
• License: MIT

Install and usage instructions live in the source repository linked above.