Menu
Takes a long-form article and repurposes it into multiple formats: tweet thread, LinkedIn post, TL;DR, and key quotes.
Install with Codex or Claude Copy this prompt, paste it into Codex, Claude, or another assistant, and let it review the skill page and install it for you.
Based on SOC occupation classification
| name | Article Repurposer |
| description | Takes a long-form article and repurposes it into multiple formats: tweet thread, LinkedIn post, TL;DR, and key quotes. |
| category | writing |
| tags | ["content","repurposing","social-media","writing"] |
| author | simplyutils |
This skill takes a long-form article, blog post, or essay and repurposes it into four ready-to-publish formats: a Twitter/X thread, a LinkedIn post, a TL;DR summary, and a set of pull-quote cards. Each format is written natively for its medium โ not just the article copy-pasted with line breaks. The output can be published immediately or used as a starting point for further editing.
Use this to maximize the reach of any long-form content you publish. A single well-written article should produce at least a week's worth of social content.
Copy this file to .agents/skills/article-repurposer/SKILL.md in your project root.
Then ask:
Provide:
Add the instructions below to your .cursorrules or paste them into the Cursor AI pane. Paste the article text.
Paste the full article and ask Codex to follow the instructions below to produce all four formats.
When asked to repurpose an article, produce all four of these formats. For each one, follow the specific instructions below.
Goal: Turn the article's core argument and key insights into an engaging multi-tweet thread that stands alone โ readers shouldn't need to read the article to get value from the thread.
Structure:
Rules:
Goal: A thoughtful, standalone post that works for a professional audience. LinkedIn rewards longer, reflective posts โ this isn't a thread, it's a mini-essay.
Structure:
Rules:
Goal: A 3โ5 bullet summary that someone could read in 30 seconds and understand the article's key takeaways. Suitable for a newsletter roundup, Slack message, or email.
Structure:
Rules:
Goal: Extract 3โ5 quotable sentences from the article that would work as standalone graphic quotes (for Twitter image cards, LinkedIn carousels, Instagram, or internal Slack pins).
Criteria for a good pull quote:
Format: List each quote verbatim from the article, followed by a one-word tone label (Surprising / Motivating / Contrarian / Tactical / Philosophical).
Deliver all four formats clearly labeled and separated:
## Twitter/X Thread
[tweets numbered 1/ through N/]
---
## LinkedIn Post
[post text]
---
## TL;DR
[summary]
---
## Pull Quotes
1. "[quote]" โ Surprising
2. "[quote]" โ Tactical
...
Input to Agent:
"Use the Article Repurposer skill on this article excerpt:
Why Most Developers Ship Bugs They Could Have Caught
The average developer spends 30% of their time debugging code โ but research from Cambridge University found that 70% of those bugs could have been caught at the design stage, before a single line of code was written. The problem isn't that developers are careless. It's that we've trained ourselves to think that thinking is less productive than coding...
[article continues โ assume full article provided]"
Output from Agent:
Twitter/X Thread
1/ 70% of bugs could be caught before you write a single line of code.
Most developers skip this step because it feels like "not working."
It's the most expensive mistake in software development. ๐งต
2/ The average developer spends 30% of their time debugging.
That's 12+ hours a week โ for a typical 40hr week.
Almost all of it is optional.
3/ The research is clear: bugs caught at the design stage cost ~10x less to fix than bugs caught in production.
But "design phase" sounds like waterfall. It isn't.
It's just: think before you type.
4/ The habit that changed how I code:
Before touching a keyboard, write down in plain English: โ What does this function receive? โ What does it return? โ What can go wrong?
Takes 3 minutes. Saves 3 hours.
5/ Read the full breakdown โ [article URL]
LinkedIn Post
70% of the bugs I've fixed in my career were preventable.
Not preventable with better tests. Not preventable with a smarter linter.
Preventable with 3 extra minutes of thinking before writing code.
Cambridge researchers found that bugs caught at the design stage cost roughly 10x less to fix than bugs caught in production. And yet most of us jump straight to the keyboard, because coding feels productive and thinking doesn't.
I spent years doing this. A deadline would hit, I'd write fast, and then spend twice as long debugging the fast code. The speed was an illusion.
The shift that worked for me: before any non-trivial function, I write three things in a comment: what it receives, what it returns, and what can go wrong. That's it. It takes minutes. It surfaces 70% of the edge cases before I have to hunt for them in production at 11pm.
What's your pre-coding ritual? I'm curious what works for other developers.
Full article: [URL]
#softwaredevelopment #programming #engineeringculture
TL;DR
Key finding: Most software bugs are introduced before coding starts โ and could be prevented with brief upfront design thinking.
- 70% of bugs could be caught at the design stage, per Cambridge research
- Developers spend 30% of their time debugging โ the majority of it avoidably
- Bugs caught in design cost ~10x less to fix than bugs caught in production
- Simple habit: write out inputs, outputs, and failure modes before coding any non-trivial function
- The feeling of "thinking isn't working" is the root cause of most debugging time
Pull Quotes
- "70% of bugs could be caught before you write a single line of code." โ Surprising
- "The speed was an illusion. I'd write fast and spend twice as long debugging the fast code." โ Contrarian
- "We've trained ourselves to think that thinking is less productive than coding." โ Philosophical
Runs a systematic checklist review on any code diff or file, covering correctness, security, performance, and readability.
Writes a high-quality CLAUDE.md, .cursorrules, or .windsurfrules file that gives a coding agent the right project context, conventions, and constraints to work effectively.
Designs an eval suite for an LLM agent or pipeline including success metrics, trajectory scoring, LLM-as-judge setup, and regression test cases.
Designs a hybrid retrieval pipeline combining dense vector search and BM25 sparse search with reciprocal rank fusion, and explains when to use each configuration.
Converts a workflow description into a LangGraph node/edge graph with typed state, conditional routing, and human-in-the-loop checkpoints.
Audits an AI application for unnecessary token spend and recommends prompt caching, model routing, and token reduction techniques to cut costs.