agentleFS
mubaidmubaid/dsh-linkedin-agent/skills/li-reply/SKILL.md

li-reply

Skill0 starsChanged 40 days agoMITView on GitHub

Handle the replies under the user's own LinkedIn posts - draft answers to every comment, sorted by which ones are worth answering. Use when the user pastes the comments on their post, says "reply to these", "handle my comments", "someone said X on my post", or is dealing with a critic or a lead in the comments.

8 other projects use a version of this, 3 of them edited.

Adopted by

Most recently taken up by

The file

We split this skill into the rules it gives an agent

1 rule. 1 is common, 0 are distinctive, and 0 are very specific to this project.

  • 1

    Common

    Most of the 4 edited versions we read keep these.

  • 0

    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. Common:

    Answer the actual question: If someone asks how, tell them how, in the reply.

    line 39

“Common” is measured: kept by at least 6 in 10 of the 4 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-reply

The reply thread under your own post is where reach is actually decided. Every reply is a second engagement event on the post, and the first hour of replies does most of the work. But the value is not equal across comments, so this skill sorts before it writes.

Input

The user pastes the comments, ideally with names and roles. Screenshots are fine. Do not scrape the thread with a browser tool.

Triage first

Sort every comment into one of five buckets and say the count out loud:

bucketwhat it iswhat it gets
LEADsomeone describing the problem you solvea real answer + a soft open door
SUBSTANCEadds data, disagrees, extendsthe longest reply on the thread
PEERa name worth being seen next toa reply that gives them something
SUPPORT"great post", 🔥, a taga like, and a 3-8 word reply at most
NOISEpitch, spam, bad faithnothing, or one line and out

Then write in that order, and stop writing when the value stops.

How to reply

  • Answer the actual question. If someone asks how, tell them how, in the reply. Do not send them to a DM to hear the answer they could have had.
  • Use their name once, at the start, and never with an exclamation mark.
  • Match their length. A two-line comment does not get a six-line reply.
  • To a critic: concede the true part first, in their words, then hold the line on the part you believe. Never delete, never get defensive, never reply twice on the same thread.
  • To a lead: answer fully in public. The open door is one sentence at the end, and it is an offer of help, not a pitch. Public value is what makes the next person DM you.
  • To a pitch in your comments: ignore it. Replying gives it reach.

Output

One block, grouped by bucket, each reply copy-ready and already humanized:

REPLIES  ·  17 comments  ·  1 LEAD, 3 SUBSTANCE, 4 PEER, 8 SUPPORT, 1 NOISE

LEAD
@Sarah Chen - "we have the same problem with proposals"
> The part that fixed it for us was pulling the pricing table out of the doc
> entirely and sending it as its own page. Cut a whole review round. Happy to
> send you the template if it is useful.

SUBSTANCE
@Marcus Webb - disagrees on cadence
> Fair, and 4x a week only worked because I had 8 months of posts behind it.
> From a standing start I would do what you said.

SUPPORT  (like these, reply to the first three)
@... "This is great" -> Thanks Dan.
...

NOISE  (1)
Skipped: an agency pitch. Replying gives it reach.

Then the gate: nothing is posted until the user says yes. They paste the replies themselves.

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.