Verified against ChatGPT · 2026-08-11
Draft LinkedIn comment replies that add something specific instead of just agreeing louder
Writes 2-3 alternative replies to a specific comment on your post or someone else's, each one built to add a genuinely new angle, disagreement, or detail — instead of the reflexive 'Great point!' that makes a comment section look active but says nothing.
The prompt
Ready to copy — highlighted parts are example details you can swap.
Help me reply to a specific comment. I don't want a generic agreement reply — LinkedIn comment sections are full of "Great point!" and "So true!" replies that add nothing and make people scroll past. I want a reply that actually contributes something. THE ORIGINAL POST OR TOPIC A founder posted about how they stopped doing weekly 1:1s and switched to async check-ins to save time. THE COMMENT I'M REPLYING TO "This only works if your team is already high-trust. For newer teams this would backfire fast." MY ACTUAL REACTION TO IT I actually disagree — we made this switch with a brand-new team and it worked because we over-communicated for the first month. RELATIONSHIP TO THIS PERSON Someone in my industry I don't know personally but whose content I respect and want to be on their radar. WHAT I KNOW THAT THEY MIGHT NOT We did this with a team of 4 new hires and it worked because we ran a 2-week overlap of both formats before dropping the weekly 1:1s entirely. REPLY RULES Write 3 reply options, each taking a genuinely different approach: one that adds a specific detail or example from my own experience that extends their point rather than just validating it, one that respectfully disagrees or adds nuance if my_reaction suggests I don't fully agree, and one that asks a real follow-up question I'd actually want the answer to (not a rhetorical or softball question). Every option must be short enough to read as a comment, not a mini-post — 1-3 sentences, never a wall of text in a reply thread. If my actual reaction is genuine agreement with nothing to add, say so plainly instead of forcing manufactured disagreement or a fake follow-up question just to hit three distinct options — tell me a short authentic agreement is the right call here and give me that instead. Match the tone to the relationship context — a reply to a stranger's post reads differently than a reply to a close colleague's, and the draft should reflect that difference rather than using one generic professional register for both. WHAT NOT TO DO Do not open any reply with "Great point" or "I love this" or "So true" — restate what specifically resonated or what you'd add instead of a content-free affirmation. Do not write a reply so long it reads as if you're trying to hijack the comment section for your own visibility — the goal is a genuine contribution, not a stealth post. OUTPUT FORMAT Up to 3 short reply options (or one, with a note explaining why fewer than 3 is the right call), each labeled with which approach it takes (add detail / respectful pushback / real question).
Customize
Optional — swap in your own details for the highlighted parts above.
Why this works
The explicit instruction to check whether genuine agreement with nothing to add is the actual honest state, and to permit a short authentic reply rather than force three distinct options, directly counters a structural bias in how language models respond to numbered-output requests: when asked for 'three reply options,' a model will generate three options even when the honest input only supports one, because satisfying the requested count is treated as part of the task unless explicitly told otherwise — this is the same failure mode that produces manufactured disagreement or forced questions in brainstorming tasks generally, and naming it directly is what actually gives the model permission to under-deliver on count when that's the more honest answer. Banning 'Great point' and similar openers by name, rather than a general 'be original' instruction, matters because those specific phrases are extremely high-frequency in professional social-media replies specifically, meaning they sit close to the top of the model's default distribution for this exact context, and only a named ban reliably routes around them rather than a synonym substitution that preserves the same content-free function. Tying tone explicitly to the relationship context prevents a second common failure: an unconstrained model tends to default to one uniformly polished, safe professional register for all replies regardless of audience, because that register is the lowest-risk choice absent other signal — providing the actual relationship forces a genuine register shift (more clipped and familiar for a colleague, more considered and credibility-building for a stranger whose attention you want) rather than one generic voice stretched across different social contexts.
What you get back
Add detail: "Fair concern — we actually did this with a brand-new 4-person team and it worked because we ran both formats in parallel for 2 weeks before fully dropping the weekly 1:1. The overlap period mattered more than team tenure did." Respectful pushback: "I'd push back slightly — trust was the outcome of overcommunicating early, not a precondition we had going in." Real question: "Curious what async check-in format you'd actually recommend for a team under 6 months old — same cadence, just written instead of live?"
Verified against
ChatGPT GPT-5.1 · 2026-08-11
Changelog
- 2026-08-11 — Initial publish, verified against ChatGPT GPT-5.1.
Need this built into your business?
If a prompt isn't enough — Google Ads management, built and maintained for you — that's Scult's day job.
EXPLORE GOOGLE ADS MANAGEMENT
