agentleFS
Sign inStart free
HemanthMChakravarthyHemanthMChakravarthy/linkedin-agent-skills/skills/li-comment/SKILL.md

li-comment

Skill2 starsChanged 43 days agoMITView on GitHub

Write a LinkedIn comment on someone else's post in the user's voice, picked from nine comment types matched to what the post actually is. Use when the user wants to comment on a post, pastes a post and says "comment on this", or asks for engagement ideas. Never writes "Great post!". Never posts the comment itself.

8 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

6 rules. 0 are common, 6 are distinctive, and 0 are very specific to this project.

  • 0

    Common

    Most of the 8 edited versions we read keep these.

  • 6

    Distinctive

    This author's own choices. Worth a look.

  • 0

    Very specific

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

  • Common
  • Distinctive
  • Very specific
  1. Distinctive:

    Read the target post twice.

    line 21
  2. Distinctive:

    No hashtags in comments. No links unless the poster asked for one.

    line 50
  3. Distinctive:

    Never pitch, never link-drop, never "check out my post on this".

    line 70
  4. Distinctive:

    Never argue to win. A counter comment that humiliates the poster costs more audience than it gains.

    line 71
  5. Distinctive:

    Never comment-bait the algorithm ("comment X and I'll DM you") under someone else's post.

    line 73
  6. Distinctive:

    If the post is rage bait, grief, or a layoff announcement, say so and offer type 9 or decline.

    line 75

“Common” is measured: kept by at least 6 in 10 of the 8 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-comment

Comments are the cheapest distribution on LinkedIn: a good one is read by the poster's entire audience. This skill writes the comment. The user posts it. It never publishes anything itself.

Before you write

  1. Load the voice profile (see li-post: ~/.agents/linkedin/voice.md first, then the Claude/Copilot fallbacks; if missing, invoke li-voice).
  2. Read the target post twice. Identify what it actually is: a story, a data drop, a hot take, a question, a milestone, advice, a contrarian claim, a how-to, or a personal moment. The type decides the comment.

The nine types

#TypeUse when the post is...The move
1Extendadvice / how-toAdd the missing step or the exception
2Counterhot take / contrarianSteelman it, then add the boundary condition
3Evidencedata drop / claimAdd your own number that confirms or complicates
4Storypersonal storyOne short parallel story, then out
5Questionany strong postAsk the question the post made you genuinely curious about
6Reframemilestone / announcementName what the milestone actually proves
7Distilllong advice postCompress their point into one quotable line, then add one nuance
8Bridgequestion postAnswer it briefly, then connect it to your own experience
9Witnesspersonal momentAcknowledge it like a person, not a brand

The loop

1. Name the type. Say which of the nine you picked and why, in one line.

2. Draft it. Rules:

  • 40-120 words. A comment is not a post.
  • Open with the substance, not the greeting. No "Great post!" and no "Thanks for sharing" anywhere in it.
  • One idea. The comment makes one point and stops.
  • Match the poster's energy but keep the user's voice.
  • No hashtags in comments. No links unless the poster asked for one.

3. Humanize it. Invoke the li-human skill before showing the draft.

4. Print the block. Copy-ready in a fenced block, then:

COMMENT READY
type:       #3 Evidence
length:     68 words
humanizer:  2 artefacts stripped, human score 88 PASS

Paste it yourself. Reply "yes" to log it.

5. Never publish. On "yes", append to ~/.agents/linkedin/log.md with the date and the commented post's author and first line.

What not to do

  • Never pitch, never link-drop, never "check out my post on this".
  • Never argue to win. A counter comment that humiliates the poster costs more audience than it gains.
  • Never comment-bait the algorithm ("comment X and I'll DM you") under someone else's post.
  • If the post is rage bait, grief, or a layoff announcement, say so and offer type 9 or decline. Not every post deserves engagement.

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.

Check your setup, free