agentleFS
HemanthMChakravarthyHemanthMChakravarthy/linkedin-agent-skills/skills/li-post/SKILL.md

li-post

Skill2 starsChanged 43 days agoMITView on GitHub

Write a LinkedIn post from a raw idea using 21 proven hook formulas, in the user's own voice, humanized so it does not read as AI. Use whenever the user wants a LinkedIn post, a hook, a draft for the feed, "post about X", "turn this into a LinkedIn post", or asks for hook options. Produces three hook options, one full draft, and a copy-ready block that is never published without an explicit yes.

9 other projects use a version of this, 9 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

7 rules. 2 are common, 1 is distinctive, and 4 are very specific to this project.

  • 2

    Common

    Most of the 9 edited versions we read keep these.

  • 1

    Distinctive

    This author's own choices. Worth a look.

  • 4

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Load the voice profile: Read the first file that exists:

    line 20
  2. Very specific:

    If none exists, invoke the li-voice skill to build one from three of the user's own posts.

    line 25
  3. Very specific:

    Load the hooks: Read hooks.json in this folder: 21 formulas with templates, filled examples, what each is for, and how each one usually …

    line 28
  4. Common:

    Demand one true thing: If the idea is thin ("post about AI"), do not pad it.

    line 31
  5. Very specific:

    1. Pick three hooks, not one: Run the idea through hooks.json and choose the three formulas that genuinely fit.

    line 54
  6. Common:

    One idea per post: If the draft has two, you have two posts.

    line 84
  7. Very specific:

    Never fabricate: No invented metrics, clients, revenue figures, or outcomes under the user's name, even as placeholders.

    line 93

“Common” is measured: kept by at least 6 in 10 of the 9 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

li-post

Turns one raw idea into a LinkedIn post that sounds like the person who posted it. This skill produces text only. It never publishes, schedules, or sends anything. The user posts the final draft themselves.

Before you write

  1. Load the voice profile. Read the first file that exists:

    • ~/.agents/linkedin/voice.md (preferred, works on every agent)
    • ~/.claude/linkedin/voice.md
    • ~/.copilot/linkedin/voice.md

    If none exists, invoke the li-voice skill to build one from three of the user's own posts. Do not skip this and do not invent a voice. A post in the wrong voice is worse than no post.

  2. Load the hooks. Read hooks.json in this folder: 21 formulas with templates, filled examples, what each is for, and how each one usually gets ruined.
  3. Demand one true thing. If the idea is thin ("post about AI"), do not pad it. Ask one batched question: what happened, to whom, and what did it cost or return. A post needs one specific, true detail. Get it first.

The shape

LinkedIn rewards dwell time, saves, and comments, in that order. So:

Line 1     The hook, alone. It must survive truncation (~140 chars on mobile).
Line 2     The payoff of line 1, not setup for line 3.
Body       Short paragraphs, 1-3 lines each, blank line between every one.
           No wall of text. The white space is the format.
The turn   One line that reframes everything above it.
Close      One specific question, OR one instruction. Never both.

Length: 900-1,300 characters is the working range for a text post. Under 400 reads as a passing thought, not a post. Over 2,000 has to earn every line, and the "see more" tap must be paid for by line 2.

The loop

1. Pick three hooks, not one. Run the idea through hooks.json and choose the three formulas that genuinely fit. Different formulas, not three variations of one. Show them as three numbered lines and say which you would ship and why, in one sentence.

2. Draft the full post on the strongest hook.

3. Humanize it. Invoke the li-human skill on the draft before showing it to the user. Every post from this skill ships humanized. That is not an optional extra step; it is the reason the draft is worth reading.

4. Print the block. Copy-ready, in a fenced code block, exactly as it should be pasted. Then, underneath:

POST READY
hook:      #17 Time Anchor
length:    1,140 characters
humanizer: 6 artefacts stripped, human score 84 PASS
post at:   Tuesday 8:15am ET (from your plan)

Reply "yes" to log it, or tell me what to change.

5. Never publish. On "yes", append the post to ~/.agents/linkedin/log.md with the date, the hook used, and the first line, so li-audit has a history to work from later. Do not invent a posting mechanism. There is none.

Rules that make the difference

  • One idea per post. If the draft has two, you have two posts. Say so.
  • Numbers over adjectives. "$4,200" beats "a lot". If the user has not given you a number, ask for one rather than writing around the hole.
  • No engagement bait. "Thoughts?" and "Agree?" are dead. The closing question must be one only this post could ask.
  • Three hashtags maximum, at the bottom, and only real categories someone actually follows.
  • No links in the post body. LinkedIn suppresses posts with outbound links. Put the link in the first comment and say so in the receipt.
  • Never fabricate. No invented metrics, clients, revenue figures, or outcomes under the user's name, even as placeholders. If a number is needed and unknown, leave {{your number}} in the draft and flag it.

Example

/li-post we built an internal tool that cut proposal time from 5 hours to 20 min
HOOKS
1. #17 Time Anchor    Writing a proposal used to take me 5 hours. It now takes 20 minutes.
2. #12 Comparison     A $12,000 proposal writer vs a weekend and a template. The weekend won.
3. #3  Mistake        For two years I charged clients for hours I was wasting on formatting.

Shipping #17: the ratio is believable and the number is yours.

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.