LinkedIn Comment Prompt
Three comments at different levels of directness, plus a verdict on whether this post is worth commenting on at all. Commenting is the cheapest way onto a prospect’s radar and the most reliably wasted: agreement adds nothing and is invisible. This prompt finds the thing the post did not say, and writes the comment that says it — which is what makes the author reply.
You write LinkedIn comments that are worth the author's reply. Agreement is not a comment.
## The post
[PASTE THE POST IN FULL, plus the author's name and role]
## Me
{{ROLE}} at {{COMPANY}}
What I know that is relevant: {{PROOF}}
Why I care about this author: {{OBJECTIVE}}
Tone: {{TONE}}
## Method
1. Decide whether this post is worth commenting on. Skip it if you have nothing to add, if the post is a truism, or if the thread is already sixty comments deep — late comments in a busy thread are invisible.
2. Find the gap: the case the post does not cover, the condition under which its advice fails, the adjacent fact the author did not mention, or the specific experience you have that tests it.
3. Write three comments:
- Additive: adds a fact or a case the post did not include.
- Questioning: asks the one question the post leaves open, genuinely.
- Dissenting: disagrees on one specific point, with the reason, and concedes the rest. Only offer this if the disagreement is real.
4. Each under 300 characters. Longer comments are collapsed and read as an attempt to write your own post under someone else's.
## Hard rules
- Never open with praise. "Great post" and "So true" are the noise this prompt exists to avoid.
- Never mention your product or your company unless the post is literally about the problem you solve, and then only once and plainly.
- No emoji, no "Couldn't agree more", no tagging people who are not part of the conversation.
- Do not restate the post's argument back to the author as if it were your own insight.
## Output format
### Worth commenting?
Yes or no, with the reason.
### Additive
Text, then (n characters).
### Questioning
Text, then (n characters).
### Dissenting
Text, then (n characters). Or: "no honest disagreement here".
### Recommended
Which one, and what a reply from the author would open up.- You want to be visible to a prospect before you message them.
- You are building an audience and have no reach of your own yet.
- You are commenting on a post you actually disagree with and want to do it without a fight.
- Comment on ten posts before you message anyone. The name being familiar is most of why the message gets read.
- Take "not worth commenting" for an answer. A comment on a truism costs you more than silence.
- Never use the dissenting option unless you mean it. A manufactured disagreement is obvious and permanent.
The output schema it demands
The single highest-leverage part of any prompt. Without a declared shape you get prose you then have to reformat; with one, you get sections you can act on — and you can tell at a glance when the model skipped something.
### Worth commenting? ### Additive (+ count) ### Questioning (+ count) ### Dissenting (+ count) ### Recommended
What it looks like filled in
Illustrative, not a promise — this is the shape of a good answer rather than a guarantee of one. The scenario is the same fictional company used across the whole toolkit.
Post by a finance-ops influencer arguing that close time is always a headcount problem. I have a customer example that contradicts this.
Worth commenting? — yes, a genuine disagreement backed by evidence.
Dissenting, 241 characters — "Disagree on this one — a customer of ours cut 3 days off a 9-day close with zero headcount change, by removing one manual reconciliation step. Headcount is sometimes the answer; entity count is the one people skip checking first."
Recommended — dissenting. It is a real disagreement backed by a specific number, not a manufactured one written for engagement.
Attach these once and stop retyping your business
This prompt works pasted into a blank chat. It works considerably better inside a Project with these attached, because then the ICP, the product and the limits are already in context and every answer inherits them.
The reference behind this prompt
Comments are not capped the way invitations are, which is exactly why they are the cheapest first touch available.
Written to a fixed structure and run through Claude and ChatGPT to confirm each one returns the output format it declares. That is not the same as a measured reply-rate benchmark, and nothing here claims one — if you run these at volume and have numbers, we want them.
Take the whole folder instead
This prompt, the files it expects and the rest of the sequence, as a zip you can drop into a Claude Project, a Cursor workspace or an agent repo. Free, no account.