agentleFS
07001ec010shnk-svg07001ec010shnk-svg/linkedin-agent-skills/skills/linkedin-post-writer/SKILL.md

linkedin-post-writer

Skill0 starsChanged 31 days agoMITView on GitHub

Generates high-performing, authoritative LinkedIn posts using 15 proven viral hook formulas and founder-mode structural frameworks.

23 other projects use a version of this.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

4 rules. 0 are common, 2 are distinctive, and 2 are very specific to this project.

  • 0

    Common

    Most of the 22 edited versions we read keep these.

  • 2

    Distinctive

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Every post generated by this skill must select and deploy one of the following 15 hook formulas within the first 140 to 210 characters (the …

    line 32
  2. Distinctive:

    When writing founder-centric content, structure the entire post body using one of these four operational frameworks:

    line 98
  3. Very specific:

    When generating a post with linkedin-post-writer, execute the following systematic procedure:

    line 133
  4. Very specific:

    Every post generated by linkedin-post-writer MUST pass through the explicit approval gate before publishing:

    line 162

“Common” is measured: kept by at least 6 in 10 of the 22 edited versions of this file we could read. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

LinkedIn Post Writer (linkedin-post-writer)

The linkedin-post-writer skill is engineered to draft viral, high-conversion LinkedIn posts for founders, technical leaders, and operators. It eliminates the blank canvas problem by applying battle-tested hook architectures and founder narrative formulas while strictly avoiding AI clichés and manipulative broetry.


When to use this skill

Ideal Use Cases:

  • Crafting original text or carousel-script LinkedIn posts from scratch.
  • Announcing major company milestones, product architecture updates, or fundraising lessons.
  • Writing founder-mode personal essays based on operational scars, hiring decisions, or pivots.
  • Creating tactical educational breakdowns for technical, B2B, or high-intent audiences.

Disambiguation ('Not for X, use Y'):

  • Not for commenting on other creators' posts -> use linkedin-comment-drafter.
  • Not for repurposing an existing YouTube video, blog post, or tweet thread -> use linkedin-repurposer.
  • Not for reverse-engineering hooks from someone else's viral post -> use linkedin-hook-extractor.
  • Not for replying to comments on your own posts -> use linkedin-reply-handler.
  • Not for building a 7-day schedule -> use linkedin-content-planner.
  • Not for scrubbing AI tells from pre-existing text -> use linkedin-humanizer.

Hook Formulas Catalog (H1 - H15)

Every post generated by this skill must select and deploy one of the following 15 hook formulas within the first 140 to 210 characters (the mobile "...see more" cutoff):

  1. H1: The Anaphora Hook
    • Repetition of identical opening words across 3 short lines, followed by a sharp contrast.
    • Example: "We tested 40 prompts. We tested 12 models. We tested 3 vector databases. Then a 5-line regex beat them all."
  2. H2: The Curiosity Gap
    • A bold, counter-intuitive premise that withholds the underlying mechanism until the reader clicks.
    • Example: "Most founders believe cold outbound email is dead. Last week our 2-person team booked 34 qualified demos with zero ad spend."
  3. H3: The Emotional Cold-Open
    • Immediate plunge into a vulnerable or crisis-level moment of high stakes.
    • Example: "At 2:40 AM, our primary production database went dark in front of 12,000 active users. That night changed our entire architectural philosophy."
  4. H4: The False-Binary Disruption
    • Presenting two conventional choices, discarding both as traps, and unveiling the hidden path.
    • Example: "You do not have to choose between burning out your engineers or missing product deadlines. Both come from broken domain boundaries."
  5. H5: The Year-Over-Year Pivot
    • A stark chronological delta illustrating undeniable compounding growth or dramatic operational turnaround.
    • Example: "November 2024: $0 revenue, 6 rejections, living off savings. November 2026: $1.4M ARR and cash-flow positive. The 4 shifts that mattered:"
  6. H6: The R.I.P. Obituary
    • Declaring a legacy standard, outdated software methodology, or conventional tactic officially dead.
    • Example: "R.I.P. the 60-minute enterprise sales demo (2010–2026). Buyers no longer want slide decks. They want interactive sandboxes."
  7. H7: The Controlled A/B
    • Contrasting amateur or average execution directly against elite top-1% execution.
    • Example: "Junior engineers write code for computers. Senior engineers write code for other engineers. Staff engineers delete code."
  8. H8: The Stacked-Triad
    • Three rapid-fire, rhythmic observations building escalating narrative pressure.
    • Example: "Unclear specs. Rushed deadlines. Fragmented Slack channels. This is how high-performing development teams quietly disintegrate."
  9. H9: The Question-Bomb
    • An unsettling diagnostic question that exposes an unspoken industry inefficiency.
    • Example: "Why do B2B companies pay $10,000 a month for marketing software when their core signup flow takes 7 clicks and fails on mobile?"
  10. H10: The Data-Reveal
    • Leading with an exact, non-rounded internal production metric or research finding.
    • Example: "We analyzed 840,000 automated background API requests last month. 71.4% of total latency occurred inside DNS lookups."
  11. H11: The Confession
    • A frank, high-accountability admission of an expensive founder mistake.
    • Example: "I wasted $65,000 building a feature our customers begged for. Nobody ever used it. Here is the validation rule I now enforce:"
  12. H12: The Prediction
    • A reasoned forecast about market restructuring over the coming 12 to 24 months.
    • Example: "By late 2027, the traditional SDR role will no longer exist in enterprise SaaS. It will be replaced by technical customer engineers."
  13. H13: The Tactical Listicle Opener
    • A precise, no-nonsense promise of curated operational assets.
    • Example: "6 open-source internal tools that replaced $2,800/month in SaaS subscriptions (all self-hostable, SOC2-ready):"
  14. H14: The Future-Vision
    • A vivid glimpse into a frictionless operational future.
    • Example: "Imagine pushing code to production with zero configuration files, zero CI pipeline flakiness, and instant rollback. Here is how we built it:"
  15. H15: The Counter-Narrative
    • Dismantling a widely shared LinkedIn platitude with raw production realities.
    • Example: "'Move fast and break things' is reckless advice for fintech infrastructure. If your ledger breaks once, your business is terminated."

Founder Mode Structural Formulas (F17 - F20)

When writing founder-centric content, structure the entire post body using one of these four operational frameworks:

F17: The Build-in-Public Architecture Reveal

  • Section 1 (Lines 1-3): Hook (H10 Data-Reveal or H14 Future-Vision) highlighting a system breakthrough or scale milestone.
  • Section 2: The initial bottleneck (what broke at previous scale).
  • Section 3: The rejected naive solutions (why simply adding bigger servers or off-the-shelf tools failed).
  • Section 4: The new architecture diagram / logic (bulleted breakdown of components, queues, and caches).
  • Section 5: Real production benchmarks (latency before vs after, cost per request before vs after).
  • Section 6: Clean call to discussion ("What is your team using for queue orchestration at scale?").

F18: The Cash Burn & Runway Reckoning

  • Section 1 (Lines 1-3): Hook (H3 Emotional Cold-Open or H5 Year-over-Year) with exact runway months or cash numbers.
  • Section 2: The realization of vanity spend (overhiring, bloated software subscriptions, unfocused marketing).
  • Section 3: The emergency surgery (the 3 hard cuts made within 48 hours).
  • Section 4: The unit economics turnaround (gross margins, CAC payback, operating efficiency).
  • Section 5: The enduring founder rule (how we evaluate capital allocation today).

F19: The Zero-to-One Customer Hustle

  • Section 1 (Lines 1-3): Hook (H2 Curiosity Gap or H9 Question-Bomb) stating zero ad spend and unscalable effort.
  • Section 2: Why automated outbound and broadcast announcements yielded zero conversions.
  • Section 3: The unscalable manual playbook (visiting offices, offering bespoke audits, building custom tools for free).
  • Section 4: The first paying champion (the pivotal conversation that unlocked product-market signal).
  • Section 5: The principle of earning early trust before trying to automate it.

F20: The Hard Non-Negotiable Operating Rule

  • Section 1 (Lines 1-3): Hook (H15 Counter-Narrative or H8 Stacked-Triad) announcing an uncompromising company policy.
  • Section 2: The painful incident that forced this rule into existence.
  • Section 3: How the rule is enforced day-to-day (specific workflows, alerts, or SLAs).
  • Section 4: The pushback faced (internal skepticism or client questions).
  • Section 5: The long-term commercial payoff (retention gains, velocity boost, team trust).

How It Works Step-by-Step

When generating a post with linkedin-post-writer, execute the following systematic procedure:

  1. Intake & Intent Discovery:
    • Extract the core topic, primary lesson, target audience, and available data/metrics.
    • Reference references/story-bank.md or references/founder-topics.md if available.
  2. Hook Selection & Tuning:
    • Select the most appropriate hook formula (H1–H15).
    • Draft 3 variations of the first 3 lines, measuring character count to ensure the cutoff occurs before 210 characters.
  3. Body Architecture & Drafting:
    • Apply a standard body progression or one of the Founder Mode formulas (F17–F20).
    • Enforce short paragraphs (1-3 lines), clear blank lines, and bulleted takeaways.
    • Maintain sweet spot character count: 900 to 1,300 characters.
  4. Voice & Polish Audit:
    • Cross-check against references/voice-rules.md.
    • Purge all banned AI vocabulary (delve, tapestry, leverage).
    • Restrict em dashes (maximum 1, preferred 0).
    • Confirm that any external URLs are scrubbed from the post body and marked for Comment #1.
  5. Approval Staging:
    • Present the post in full to the user with hook metadata and character counts.
    • Stage the post via PubloraClient and present the approval ticket.

Approval Gate

Every post generated by linkedin-post-writer MUST pass through the explicit approval gate before publishing:

[STAGED FOR APPROVAL]
Action: LIVE POST
Target Platform ID: urn:li:person:abcdef1234
Post Length: 1,142 characters (within optimal 900-1,300 range)
Hook Applied: H10 (Data-Reveal)
First Comment Link: https://github.com/example/open-source-tool

Approval Token: 8f4b1e9c-2b12-4d7a-a621-998877665544
Status: HELD IN APPROVAL GATE. No network request made.
Type 'APPROVE 8f4b1e9c-2b12-4d7a-a621-998877665544' to dispatch to LinkedIn.

The system will never automatically dispatch or publish without human authorization.


Examples

Example Prompt 1: Technical Architecture Post

User Prompt: "Write a post about how we replaced our slow microservices architecture with a modular monolith and cut our cloud bill in half."

We analyzed 1,400,000 API requests hitting our Kubernetes cluster last month.
Over 45% of total request latency wasn't business logic.
It was internal network hops between 14 microservices.

Here is what happened when we killed the microservices and migrated back to a modular monolith:

1. Infrastructure Costs:
We decommissioned 8 AWS instances and consolidated onto 2 dedicated servers. Cloud bill dropped from $8,400/mo to $3,200/mo overnight.

2. P99 Latency:
Inter-service HTTP serialization disappeared. End-to-end response times plummeted from 380ms to 24ms.

3. Developer Velocity:
Instead of debugging distributed tracing and Helm charts across 14 repositories, engineers run one local command and have the full application running in 6 seconds.

Microservices solve organizational scaling problems for teams of 500 engineers.
For a team of 15, they are an unforced error.

Link to our technical RFC and migration roadmap in the first comment.

Example Prompt 2: Founder Mode Cash Burn Post

User Prompt: "Write a post using formula F18 about the time we realized we had 60 days of runway left and how we fixed our unit economics."

September 2024: 62 days of cash left in the bank account.
8 full-time salaries relying on our payroll clearing every two weeks.
I sat in my car after an investor call and realized nobody was coming to save us.

We were spending $28,000/month on marketing agencies, outbound lead lists, and shiny SaaS tools.
Our revenue? Just $14,000/month.

We made 3 emergency decisions within 48 hours:

1. We fired the marketing agency.
Founders must do the initial prospecting until customer acquisition is predictable.

2. We canceled 19 SaaS subscriptions.
If a software tool didn't directly write to our production database or process customer payments, it got cut.

3. We 3x'd our prices for all new inbound leads.
We stopped pitching $49/mo plans and offered a $500/mo managed onboarding package.

Within 45 days, our monthly burn dropped by 60%, and our revenue climbed to $38,000/mo.
We reached cash-flow breakeven without giving away another percent of equity.

Growth doesn't cure bad unit economics. Discipline does.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.