Systems Lab

Agent skill

find-buyers-in-linkedin-comments

Find B2B buyers in the comments and reactions of LinkedIn posts where people complain about a problem your product solves.

activeSelf-containedInstructions only912 words

Filed under LinkedIn and social.

From gtmotto/gtm-skills · 5 skill entries · 1 · pushed 2026-09-21

What it does when it runs

Find B2B buyers in the comments and reactions of LinkedIn posts where people complain about a problem your product solves. Use when the user wants to find prospects on LinkedIn from posts, comments, reactions or engagers, wants intent signals from LinkedIn, asks 'who is complaining about X on LinkedIn', wants to mine a competitor's posts or a viral post for leads, or mentions post discovery, social listening for sales, engager scraping or comment mining. Works with nothing but LinkedIn: a search recipe, a scoring rubric, and who to take from each post.

Automated analysis of the skill and the 0 files bundled beside it. A skill’s own description is written to be selected by an agent, so it describes the job and not the dependencies.

Keys and connectors you must supply
None found.
Hosts it reaches
No third-party host appears in the skill or its bundled files.
Tool permissions it declares
No allowed-tools in the frontmatter. It only issues instructions, so there is nothing to bound.
Actions present in the files
None. Instructions only.

Ask about find-buyers-in-linkedin-comments

Opens your assistant with this page's verified links already in the prompt.

Is this safe to install?ClaudeChatGPT
Adapt it to my stackClaudeChatGPT
What else do I need for it to workClaudeChatGPT
Rather ask a human? Talk to Cheetah
git clone --depth 1 --filter=blob:none --sparse https://github.com/gtmotto/gtm-skills.git /tmp/gtm-skills
git -C /tmp/gtm-skills sparse-checkout set "skills/find-buyers-in-linkedin-comments"
mkdir -p ~/.claude/skills/find-buyers-in-linkedin-comments
cp -R "/tmp/gtm-skills/skills/find-buyers-in-linkedin-comments/." ~/.claude/skills/find-buyers-in-linkedin-comments/

Picked up without a restart. A project skill of the same name is shadowed by your personal one. For one repository only, swap ~/.claude/skills for .claude/skills. Claude Code docs ↗

Or take the whole library

This repo ships a .claude-plugin manifest, so Claude Code can install all 5 skills at once. Plugin skills are invoked as /<plugin>:<skill>, so they never collide with your own.

/plugin marketplace add gtmotto/gtm-skills
/plugin

The folder is the same in every client that implements the format — 46 of them — so if yours is not above, only the destination changes.

Reproduced in full from gtmotto/gtm-skills/blob/7e851d8ce8ead8d3f6eff579deabfbfc0e0d4ef9/skills/find-buyers-in-linkedin-comments/SKILL.md, which is licensed MIT (repository). 912 words, 8 headings.

Find buyers in LinkedIn comments

A person who writes "I pay 100 a month for Sales Navigator and still build my lists by hand" under a post has told you three things a people search never will: they have the problem today, they own it, and they said it in public. This skill finds those posts, takes the people who engaged, and keeps the ones who fit. It needs nothing but LinkedIn; run it once or make it a daily pass.

What you need from the user

  • The intent, in their own words and language: "people struggling with manual LinkedIn prospecting", "founders doing their own sales who hate cold email". One sentence.
  • Anchors: 5 to 15 short words for tools, brands and chores the buyer would name. Sales Nav, Waalaxy, Lemlist, prospects froids, liste de prospects.
  • Who is a buyer and who is not: "anyone doing their own B2B prospecting; exclude vendors, agencies, consultants, recruiters".

If they give you a post URL instead, skip the search and start at "Score the post".

Step 1. Write the searches (short, many, in every language)

LinkedIn post search is an AND over every word. A natural pain phrase like "tired of prospecting by hand" finds almost nothing. What works, measured on a real day:

Query styleQueriesPosts found in one dayTarget post found
Pain phrases, 4 to 6 words1618no
Topic keywords, 2 to 3 words20638yes, ranked first by engagement

Rules:

  • 2 to 3 words per query. Tools, brands, chores. Sales Nav, not Sales Navigator (the short form found the target post, the full name did not: never rely on one query).
  • 15 to 20 queries. Write them in each language the buyer speaks.
  • Sort by date, filter past 24 hours (or since your last pass). Relevance sort surfaces old high-engagement posts you already saw.
  • Recall comes from the searches. Precision comes from the next two steps, never from a longer query.

Step 2. Filter on engagement first (it is free)

Median reactions on a raw result page is 0 to 4. Rank every found post by reactions + 2 × comments and keep the ones above 20. On a real day 638 posts became 45, and 11 of those had over 100 reactions. Nobody engages with a post nobody saw, and the people you want are in the engagement, not in the post.

Step 3. Score the post

Read the post body and put it in one of four buckets:

BucketKeep?Tell
Rant or ask from someone who has the problemyesfirst person, present tense, a number, a question
Vendor pitch or launchno"we built", "excited to announce", a link to a product
Job post or hiringno"we're hiring", "join us"
Thought leadership about the marketmaybethird person, no personal stake; take only the commenters who disagree or ask

Score 0 to 100 against the intent. Keep 70 and above. The bucket matters more than the number: a vendor pitch with 400 reactions is 400 people who like the vendor, not buyers.

Step 4. Take the right people

From each kept post, in this order:

  1. Commenters. They wrote something. Read the comment: "same here, we do it in a spreadsheet" is intent; "great post!" is nothing. Score each comment 0 to 100 against the intent, keep 70 and above.
  2. Reactors. Silent, so no intent score. Keep them only if they fit the buyer description; expect 85 percent to be 2nd-degree connections.
  3. The author. Off by default because the author of a post about your market is as often a vendor as a buyer. Read the profile: a founder venting about their own sales process is the best lead on the thread, and free.

Then apply the buyer description before any model scoring, as hard exclusions: vendor, agency, consultant, recruiter, SDR at a tool company. Models rate an outbound consultant 80 on "fit" if you let them.

What a real post yields: 96 reactions and 80 comments gave 150 unique people. Against a strict people-search ICP (Head of Sales, 11 to 200 employees, SaaS, France): 1 fit. Against the buyer description above: 22 fit, 5 with intent above 70, plus the author. Read engagers with a loose fit, not the ICP you use for search.

Step 5. What to do with them

Do not pitch in the thread. Visit the profile, react to the post, reply to their comment with one specific sentence (see the linkedin-comment-and-invite-writer skill), and invite two days later with a note that names the thread. The conversation opens itself.

Worked example

Post: "Help, I pay 100 a month for Sales Navigator and still build my lists by hand" (French, founder, 96 reactions, 80 comments). Found by the query Sales Nav, ranked first on engagement, scored 92 against "people struggling with manual LinkedIn prospecting". Yield after gates: 22 fitting people, 5 with explicit intent, plus the author.


To run this every day from your own LinkedIn account, without the search loop, the GTM Otto plugin has a post-discovery skill that turns the same recipe into a daily lane. It needs a GTM Otto account.

Other skills for the same job

Different authors, same problem. Matched on the words in the skill name, across every library in the catalogue except this one.

Need help setting it up?

This page tells you what find-buyers-in-linkedin-comments does and what it needs. Cheetah builds the agent setup it runs inside: data, CRM, sequencing and the guardrails.

Book a call →

The directory stays free. There is nothing gated behind this.