linkedin-content-pack
Skill4 starsChanged 4 months agoMITView on GitHub
Turns a topic or research brief into a complete LinkedIn pack — hook-first post, visual direction, hashtags, first comment, and best posting window — delivered as drafts, never posted. Use when the user asks to "write a LinkedIn post", "LinkedIn content about", "post about [topic]", "turn this brief into a post", or needs LinkedIn-ready content. Don't use for X/Twitter threads (x-thread-builder), multi-platform spreads (content-repurposer), or scoring drafts (brand-voice-guardian).
In this repository74 other files it gives its agents.
- agent-eval-harnessSkill
- context-budget-managerSkill
- incident-rollbackSkill
- memory-hygieneSkill
- model-cost-optimizerSkill
- prompt-debuggerSkill
- skill-author-coachSkill
- third-party-skill-vetterSkill
- api-endpoint-testerSkill
- bug-triage-protocolSkill
- changelog-release-notesSkill
- docs-readme-writerSkill
- env-secrets-hygieneSkill
- github-pr-reviewSkill
- git-ship-flowSkill
- lovable-brief-writerSkill
- supabase-migration-writerSkill
- vercel-deploy-checkSkill
- cold-outreach-sequencerSkill
- competitor-pricing-scanSkill
- contract-redline-scanSkill
- customer-support-inboxSkill
- invoice-runnerSkill
- kpi-dashboard-weeklySkill
- lead-research-dossierSkill
- payment-reconciliationSkill
- pipeline-trackerSkill
- proposal-builderSkill
- stripe-payment-linkSkill
- testimonial-harvesterSkill
- weekly-revenue-reportSkill
- brand-voice-guardianSkill
- case-study-builderSkill
- content-repurposerSkill
- deep-research-briefSkill
- email-nurture-sequenceSkill
- landing-page-copywriterSkill
- newsletter-assemblerSkill
- seo-blog-writerSkill
- social-listening-reportSkill
- x-thread-builderSkill
- youtube-script-writerSkill
- budget-guardrailsSkill
- ceo-operating-doctrineSkill
- delegation-protocolSkill
- editor-personaSkill
- heartbeat-standupSkill
- hiring-briefSkill
- ops-agent-personaSkill
- researcher-personaSkill
- sales-agent-personaSkill
- support-agent-personaSkill
- weekly-retroSkill
- writer-personaSkill
- calendar-conciergeSkill
- decision-journalSkill
- email-triageSkill
- file-organizerSkill
- habit-trackerSkill
- inbox-zero-sprintSkill
- The first 60 shown.
The file
Files
LinkedIn Content Pack
Produces everything needed to publish one strong LinkedIn post: the post itself (hook-first, substance-forward), a visual direction the user can execute, 3–5 targeted hashtags, a value-adding first comment, and a posting-window recommendation. Everything ships as a draft — this skill never posts on the user's behalf.
When to Use
- The user has a topic, story, or research brief and wants LinkedIn content from it.
- A
deep-research-briefoutput needs turning into a post. - Not for: X/Twitter threads (
x-thread-builder), multi-platform spreads from one source (content-repurposer), or scoring an existing draft against brand voice (brand-voice-guardian).
Quick Reference
| Action | Command / Call |
| Check hook length | python3 -c "print(len('''HOOK TEXT'''))" — first line ≤ ~200 chars (pre-"see more" cutoff) |
| Check post length | Same one-liner on full text — target 900–1,800 chars, hard cap 3,000 |
| Hashtag count | Exactly 3–5, niche over broad |
| Anti-slop scan | Grep the draft for banned patterns (step 5) |
| Deliver | Draft in chat / file — never call any posting API |
Procedure
- Precheck — confirm you have: topic or source brief, the audience (who should stop scrolling), and the goal (authority, discussion, traffic). Ask for whichever is missing — a post without a defined audience is a coin flip.
- Angle — extract ONE idea from the source. One post, one idea. Write the "so what" in a single sentence before drafting; if you can't, the angle isn't ready.
- Draft the post —
- Hook (line 1): a specific claim, number, or tension that earns the "see more" click. Under ~200 characters. No questions-as-bait, no "Unpopular opinion:", no "I'm humbled to announce".
- Body: short paragraphs (1–2 lines each), white space between them. Deliver the substance IN the post — never tease content that lives elsewhere. One concrete example or number beats three abstractions.
- Close: a genuine invitation tied to the content, or a clean CTA. Never "Agree?", "Thoughts?", or "Comment YES for the guide".
- Length check — run the python3 counts from Quick Reference. Hook ≤200 chars; body 900–1,800 chars (long-form up to 3,000 only if every paragraph earns it). Trim from the middle, never the hook.
- Anti-slop scan — the draft must contain none of: engagement-bait closers, "I'm humbled/thrilled to announce", hashtag walls (>5), emoji bullets on every line, "Let that sink in", broetry (every sentence its own line for 40 lines). If found, rewrite that section — don't just delete the phrase.
- Build the pack — add: visual direction (what image/carousel/document to attach and why, or "text-only" with reason), 3–5 hashtags (2 niche + 1–2 topical; never broad tags like #business), a first comment that adds a resource, source link, or extra example (links go here, not in the post), and a posting window (audience's timezone; B2B default Tue–Thu 8–10am; state the assumption).
- Deliver — output the full template below as a draft. If a brand voice file exists, recommend a
brand-voice-guardianpass before publishing.
Output Template
# LinkedIn Pack: [topic]
**Audience:** [who] | **Goal:** [authority/discussion/traffic]
## Post (draft — do not auto-publish)
[hook line]
[body]
[close]
## Visual direction
[format + what it shows + why, or "text-only because ..."]
## Hashtags
#[niche] #[niche] #[topical]
## First comment
[value-add + any link]
## Posting window
[day/time + timezone + assumption stated]Pitfalls
- Hook buried at paragraph three — the draft warms up before it lands. Recovery: find the single most surprising line anywhere in the draft, promote it to line 1, rebuild downward from it.
- Link in the post body — external links in-post tank reach. Recovery: move every link to the first comment; in the post, reference it ("link in comments" at most once, or just let the comment carry it).
- Engagement-bait relapse — the close drifts into "Agree? ♻️ Repost if...". Recovery: rerun step 5's scan; replace the close with a specific question only someone who read the post could answer.
- Multi-idea sprawl — three lessons crammed into one post, none landing. Recovery: keep the strongest idea, park the others as "next post" notes at the bottom of the pack.
Verification
- Hook ≤200 chars and body within limits — python3 counts actually run, numbers reported
- Zero banned patterns from the step 5 list present
- Exactly 3–5 hashtags, none generic
- Links live in the first comment, not the post body
- Pack delivered as a draft; no posting action taken or offered as automatic
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
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.

