--- name: linkedin description: | Write LinkedIn posts and articles that earn attention without performing for it. Use for professional social posts, thought-leadership pieces, and career or company announcements. Enforces the EW anti-AI rules and the writer's voice profile. license: MIT metadata: version: "0.3.0" ---

# EW Skill — LinkedIn

**Dependency chain. Read these in order before writing anything:**

1. `core/anti-ai-rules.md` (read fully; Section 0 sets the standard, 0.1 sets precedence, 0.2 forbids fabrication, 9 and 10 stop over-correction) 2. `core/ai_slop_commandments.md` (mechanism reference and diagnostic checklist) 3. `core/voice-profile.md` (this writer's voice fingerprint; if `Completed: No`, stop and run onboarding) 4. Any `.md` files the user has dropped in `references/`

This chain applies on direct invocation (`/ew:linkedin`) exactly as it does on dispatch through `/ew`. Direct invocation skips routing, not constraints.

---

## WHAT THIS SKILL IS FOR

LinkedIn posts that earn attention without performing for it. Thought leadership that has actual thoughts. Professional writing that doesn't sound like a press release from a committee.

LinkedIn is the most AI-contaminated writing surface on the internet. The bar for standing out is not high — it's just not zero.

---

## THE PLATFORM REALITY

LinkedIn's algorithm rewards engagement velocity — comments and shares in the first 60 minutes matter most. But chasing the algorithm at the cost of your voice is the fastest way to sound like everyone else. The correct approach: write something worth reading, format it for the platform, ship it consistently. The algorithm follows quality over time.

**What LinkedIn readers do:** They scan. They stop at things that feel like they apply to their situation right now. They share things that make them look smart or thoughtful to their network. Write with that behavior in mind, not against it.

---

## HOOK ARCHITECTURE

The first 2–3 lines are what show before the "see more" cutoff. These are not the intro — they are the entire post compressed into its most compelling version. If the hook doesn't earn the click, nothing below it matters.

**Hook formulas that work on LinkedIn:**

**The Counterintuitive Statement:** "The best managers I've worked with were the ones who seemed least busy."

**The Specific Observation:** "I've read 40+ cold emails this month. Exactly 3 opened with something other than a compliment about my work."

**The Uncomfortable Admission:** "I spent 6 months optimizing a product feature nobody wanted. Here's what I missed."

**The Pattern Interrupt:** "Everyone says to post consistently. Nobody talks about the 18 months before anyone notices."

**What kills the hook on LinkedIn:** - Starting with your name or job title - Starting with "I'm excited to share..." - Starting with a compliment to someone else ("So proud of [Name]...") - Starting with a question that's too vague to create urgency - Starting with "As a [title], I've learned that..."

---

## POST STRUCTURE

LinkedIn prose needs white space to be readable on mobile. This is not optional — walls of text get scrolled past regardless of the quality inside them.

**The standard structure:**

**Line 1–2:** Hook (hard stop before the "see more" cutoff) **Line 3–4:** The setup — what situation or observation you're building from **Body:** The core idea, broken into single-sentence or two-sentence paragraphs with line breaks between each **Final line:** The sharpest point, a specific takeaway, or a genuine question that invites a specific type of response

**Line break rules:** - One idea per paragraph - Max 2–3 sentences before a break - No bullet walls — if you use bullets, max 4, keep them short, don't bold the first word of each one - Em-dashes sparingly — 1 per post maximum

---

## TONE: PROFESSIONAL AUTHENTICITY

LinkedIn has two failure modes: corporate stiffness and performed vulnerability. Both are unreadable.

**Corporate stiffness sounds like:** "Leveraging synergistic frameworks to drive stakeholder alignment across cross-functional teams has been a key component of our success trajectory." This is how no human speaks. Delete it.

**Performed vulnerability sounds like:** "I'll be honest — I've struggled with this too. And if you're reading this, I bet you have as well. We're all in this together." This is emotional manipulation dressed as relatability. It's the false vulnerability pattern from `anti-ai-rules.md`. Don't do it.

**Professional authenticity sounds like:** A specific thing you observed, learned, or got wrong, stated plainly, with enough detail that it's clearly real and not a writing exercise.

The test: could this sentence have been written by a thousand different people on LinkedIn? If yes, rewrite it until it couldn't.

---

## THE INSIGHT REQUIREMENT

Every LinkedIn post must have a point that is non-obvious enough to be worth the reader's 45 seconds.

**The three-second test:** Read your opening line. Does the reader's next thought need to be "I want to know more about this" — or could they reasonably think "I already know this" or "so what"? If it's the latter, the insight isn't sharp enough.

**Specificity over generality:** ❌ "Leadership is about showing up for your team." ✅ "The best thing I did for my team last quarter was cancel the weekly status meeting and replace it with a shared async doc. Morale went up. I heard more, not less."

Specific experiences carry proof. Platitudes carry nothing.

---

## WHAT LINKEDIN IS NOT FOR

- Thought experiments with no real position ("Curious what you all think about AI... the answer might surprise you!") - Humblebrag announcements dressed as lessons ("Humbled and grateful to have been recognized by Forbes 30 Under 30... here's what I learned from the process") - Corporate announcements that belong in a press release - Hot takes without substance — having an opinion is not the same as having a point

---

## CALL TO ACTION

Not every post needs one. When you use one, make it specific enough to be answerable.

❌ "What do you think? Share your thoughts below!" ✅ "What's the one system that's actually worked for your team?"

A vague CTA signals the writer doesn't know what they want from the reader. A specific CTA signals the writer knows exactly what conversation they want to start.

---

## POST-GENERATION REVIEW

After producing any draft, stop. Re-read `core/anti-ai-rules.md` Sections 1–7 with the draft in front of you. Go line by line through the pre-publish checklist in Section 7. Fix every failure before presenting the output.

LinkedIn is the most AI-contaminated surface this system writes for. The review pass is not optional here — it is where the differentiation actually happens.

## PRE-POST CHECKLIST

- [ ] Does the hook work as a standalone — does it earn the "see more" click? - [ ] Is there white space — is this readable on mobile? - [ ] Is there a specific insight, not just a topic? - [ ] Is the tone authentic or is it corporate/performed? - [ ] Does the post end on the sharpest point? - [ ] Have you run Sections 1–7 of `core/anti-ai-rules.md`? - [ ] Is there any "I'm excited/humbled/grateful" opening? Delete it. - [ ] Is there any bold-first bullet formatting? Restructure it. - [ ] Could this sentence have been written by anyone? Rewrite it until it couldn't.