mcpbeat Sign in

Linkedin Style Agent Skill

Use when drafting or editing a LinkedIn post in the house voice.

6k tokens
context cost
the whole folder, loaded on every use
6
files
instructions only
0
copies elsewhere
how many repositories repackaged it
17
stars on the repo
on the repository, not the skill itself

Install

one command, takes just this skill from the repository
npx skills add https://github.com/vercel-labs/eve-content-agent-template --skill linkedin-style

What comes with it

19 665 bytes besides the instruction
references/ai-phrases-to-avoid.md
references/banned-words.json
references/best-practices.md
references/plain-english-alternatives.md
references/post-specs.md

The instruction itself

4 sections, as written by the author

LinkedIn voice & style

When writing or editing a LinkedIn post:

  • Hook first. Only ~200 characters show on mobile before "…see more" — the first line

decides whether anyone expands the post. Open on the surprising stat, the bold claim, an open

loop, or a one-line story setup. No throat-clearing, no "A few thoughts on…".

  • One idea per post. If you have two, write two posts. How-tos and concrete lists land best.
  • Short lines and single-sentence paragraphs, with white space between them — it's read on a

phone, and the gaps keep it scannable.

  • Concrete and specific: a number, a result, a before/after, a moment. No vague inspiration.
  • Deliver the value in the feed. Self-contained text is what people engage with — don't make

the reader leave to get the point.

  • End with a genuine question or a clear takeaway, not a "thoughts? 👇" prompt. A real question

earns comments, and comments move the post further than likes.

  • At most one emoji, and only if it earns its place. 3–5 relevant hashtags max, grouped at the

end, CamelCase and specific (#BusinessWriting, not #Business) — they're topic signals, not

a discovery feed.

  • No "I'm humbled to announce", no engagement-bait, no "Agree?" closers or reaction-poll chains.
  • Links suppress reach. When a post needs a link, put it in the first comment and say "link

in the comments" — don't put it in the body.

Structure

  • Hook — the surprising claim or the moment, line one (before the fold).
  • Turn — the tension, or the thing most people get wrong.
  • Payoff — what you learned, with a concrete detail.
  • Close — one question or one takeaway.

References

  • references/best-practices.md — researched tactics as a pre-flight checklist (hook, links,

hashtags, formats).

  • references/post-specs.md — character limits and document/carousel specs for quick lookup.
  • references/banned-words.json — words to avoid. Read it once up front for awareness, then run

the lint_against_style tool (surface linkedin) on your draft to check it against this file

before proposing the draft to the writer.

Shared references

Shared global references for all skills in the content agent. These apply regardless of the

content type.

  • references/ai-phrases-to-avoid.md — AI-tell words, phrases, and punctuation to avoid.
  • references/plain-english-alternatives.md — plain-English swaps for bloated or vague wording.

How to use it

Copy the folder

Take vercel-labs/eve-content-agent-template-linkedin-style from the repository into ~/.claude/skills for personal use, or into .claude/skills inside a project.

Check the name does not clash

The agent identifies a skill by the name field in its header. Two skills with the same name cannot sit side by side — one of them will be ignored.