agentleFS
matteotittamatteotitta/claude-marketing-linkedin/skills/linkedin-expert-posts/SKILL.md

linkedin-expert-posts

Skill0 starsChanged 31 days agoMITView on GitHub

Writes authority LinkedIn posts in the author''s voice: contrarian takes, frameworks, industry analysis, expert breakdowns and predictions. Each post puts proof in the first lines, makes one point, and ends on one belief you want the buyer to hold, so it builds the trust that turns into messages and calls. Triggers: "expert post", "thought leadership post", "hot take", "framework post", "industry take", "contrarian take", "what most people get wrong about". Not for personal stories (use linkedin-personal-posts).

The file

We split this skill into the rules it gives an agent

3 rules. About 0 are common, 2 are distinctive, and 1 is very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈2

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 1

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Write the posts that make a buyer think "this person knows what they are doing": an opinion backed by experience, a framework, a reading of …

    line 10
  2. Very specific:

    Not for personal stories (/linkedin-personal-posts), carousels (/linkedin-carousels) or comments (/linkedin-comment).

    line 17
  3. Distinctive:

    Links: If the post needs a link, put it in the post.

    line 74

≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “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 expert posts

Write the posts that make a buyer think "this person knows what they are doing": an opinion backed by experience, a framework, a reading of what is changing in the market. These sit between personal stories (which build likeability) and offers (which ask for something).

When to use it

  • The author has a real point of view on a topic the buyer cares about.
  • You want posts people save and forward to their team.

Not for personal stories (/linkedin-personal-posts), carousels (/linkedin-carousels) or comments (/linkedin-comment).

Inputs

InputRequired?Where it comes from
Topic and the author's position on itRequiredThe user or context/expert-pov/
Who the buyer isRequiredcontext/icp/
VoiceRequiredthe tone-of-voice file in context/brand/
Proof (results, clients, data, years doing it)Recommendedcontext/expert-pov/, context/win-loss/ or the user
PositioningRecommendedcontext/positioning/
A hookOptional/linkedin-hooks

Check before writing: is there a real position here, not a summary of the obvious? Does the author have standing to say it? If a statistic is used, is the source real? If the expert POV file is missing, ask the user what they believe that most of their peers do not. Note what was missing in the output.

Agency users: swap context/ for clients/<client>/context/.

Post types

TypeUse it whenShape
Contrarian takeChallenging accepted wisdom"Everyone says X. Here is what I have seen instead."
FrameworkExplaining a system you use"The 3 checks I run before..."
Industry analysisReading data or a market shift"This changed. Here is what it means for you."
Expert breakdownMaking something hard simpleStep by step, one idea per step
PredictionForward-looking opinion"This will stop working. This replaces it."

Structure

Hook (1-2 lines)       proof + the reader named
Position (1-3 lines)   the author's actual view, stated plainly
Body (2-3 points)      specific problems and what they cost, from experience
Ending (2-3 lines)     lands one belief the reader should now hold
Call to action         only if this is the week's offer post; otherwise end on the belief

Write in this order: hook, then ending, then body. Decide where the post lands before writing the middle, so the body is a bridge to that belief. It still reads top to bottom.

Steps

  1. Pick the type from the table. Produces: the post type and the one point in 10 words or fewer.
  2. Choose the proof line. The most relevant proof, not the most impressive. Produces: the line that goes in the hook.
  3. Name the belief. One belief the reader should hold after reading (for example: "a positioning problem looks like a pipeline problem"). Produces: the ending's target.
  4. Write the hook with the proof and the reader in it. Produces: the first two lines.
  5. Write the ending that lands the belief. Produces: 2 to 3 closing lines.
  6. Write the body as the author's own experience ("on the last three projects I..."), naming what goes wrong and what it costs. Insight, not a manual. Produces: the full draft.
  7. Run the checks below and trim 15 to 20 percent. Produces: the final post.

Checks

  • Expert test. Would someone who has never heard of the author conclude they are an expert from this post alone? Proof sits in the first two lines.
  • One thing. Zoom in on one point. If there are 4 or more numbered steps, each could be its own post.
  • Insight, not manual. The reader should think "that makes sense", not "now I can do it without help".
  • Broad enough. It speaks to the whole ICP, not one client's edge case.
  • No false contrasts. Scan every line for "It's not X, it's Y", "Stop X. Start Y." and two-sentence pairs where the second reverses the first. Rewrite as a plain positive statement.
  • No bait. No "What's your take?", "Agree?" or "Thoughts?" endings.
  • Links. If the post needs a link, put it in the post. Never "link in the comments".

Mistakes to avoid

  • Invented statistics or studies. Use [data point needed].
  • Using "a client I worked with" when the user gave no such example.
  • A framework that is all structure and no view.
  • Product mentions in a post meant to build authority.
  • Hedging the position until it says nothing.

Output

Return the post ready to paste, plus: post type, the belief it ends on, the proof used, a visual suggestion (1080 x 1350 portrait if an image helps), and any placeholders to fill. For a batch, save to outputs/social/linkedin/MMYY-expert-posts.md.

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.