linkedin-comment-drafter
Skill0 starsChanged 31 days agoMITView on GitHub
Drafts high-authority, practitioner-grade comments on any LinkedIn post using Additive, Contrarian, and Question-forward frameworks.
6 other projects use a version of this.
Adopted by
Most recently taken up by
- uplika/uplika-plugintheir own version, first seen Nov 8, 2026
- CodesbyFebin/linkedin-managertheir own version, first seen Oct 20, 2026
- AIProUK/linkedin-claude-skillstheir own version, first seen Oct 20, 2026
- sylvan-io/linkedin-skillstheir own version, first seen Oct 11, 2026
- subu53/linkedin_exponentialtheir own version, first seen Oct 8, 2026
- sergebulaev/linkedin-skillstheir own version, first seen May 19, 2026
Other versions6 projects use a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 6
- sergebulaev/linkedin-skills★ 5.4ksame name
- uplika/uplika-plugin★ 3same name
- AIProUK/linkedin-claude-skills★ 1same name
- sylvan-io/linkedin-skills★ 1same name
- CodesbyFebin/linkedin-manager★ 0same name
- subu53/linkedin_exponential★ 0same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository14 other files it gives its agents.
- AGENTS.mdAGENTS.md
- CLAUDE.mdCLAUDE.md
- linkedin-marketingSkill
- linkedin-content-plannerSkill
- linkedin-employee-advocacySkill
- linkedin-engager-analyticsSkill
- linkedin-hook-extractorSkill
- linkedin-humanizerSkill
- linkedin-interviewerSkill
- linkedin-post-writerSkill
- linkedin-profile-optimizerSkill
- linkedin-reply-handlerSkill
- linkedin-repurposerSkill
- linkedin-thread-monitorSkill
The file
We split this skill into the rules it gives an agent
9 rules. 0 are common, 6 are distinctive, and 3 are very specific to this project.
0
Common
Most of the 6 edited versions we read keep these.
6
Distinctive
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 56
Objective: Ask a piercing, diagnostic question that compels the post author and other senior practitioners to respond directly to your …
- Distinctive: line 82
Banned AI Vocabulary: Do not use *delve*, *tapestry*, *leverage*, *game-changer*, *paradigm*, or *synergy*.
- Very specific: line 88
When executing linkedin-comment-drafter, follow these sequential steps:
- Very specific: line 92
If a URL is supplied, run lib.url_parser.parse_post_url(url) to validate the URN and ensure it targets a valid activity, share, or ugcPost.
- Distinctive: line 101
Verify that each option incorporates concrete terminology, avoids banned AI words, and maintains professional brevity.
- Very specific: line 108
Wait for explicit user confirmation.
- Distinctive: line 114
Before submitting a drafted comment to the approval gate, verify:
- Distinctive: line 126
No comment is ever published directly to LinkedIn without user confirmation:
- Distinctive: line 157
The No-Self-Promotion Rule: Never link your own tool or drop a sales pitch in the comment.
“Common” is measured: kept by at least 6 in 10 of the 6 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.
Files
LinkedIn Comment Drafter (linkedin-comment-drafter)
The linkedin-comment-drafter skill generates strategic, high-visibility comments on third-party LinkedIn posts from a post URL or raw text snippet. It is designed to capture the attention of post authors, target prospects, and industry peers without resorting to sycophantic praise, corporate platitudes, or unearned flattery.
A great comment on LinkedIn acts as a mini-billboard: it routinely outperforms the original post in profile visits if it earns the pinned "Top Comment" slot.
When to use this skill
Ideal Use Cases:
- Executing an outbound engagement strategy targeting industry creators, prospective buyers, or potential investors.
- Building domain authority by commenting early on trending posts in your industry niche.
- Driving high-intent profile visits from commenters and readers who view top comments on viral posts.
- Crafting well-researched, substantive perspectives on industry news, benchmark reports, or controversial debates.
- Engaging with prospect posts prior to initiating cold outbound messages.
Disambiguation ('Not for X, use Y'):
- Not for replying to comments people left on YOUR own posts -> use
linkedin-reply-handler. - Not for writing a brand new standalone post -> use
linkedin-post-writer. - Not for analyzing the profile of people who liked or commented -> use
linkedin-engager-analytics. - Not for tracking ongoing comment threads over 24 hours -> use
linkedin-thread-monitor. - Not for rewriting existing blog posts into LinkedIn updates -> use
linkedin-repurposer. - Not for cleaning AI jargon from an existing post draft -> use
linkedin-humanizer.
The Three Core Comment Frameworks
The skill supports three distinct engagement styles, each calibrated for specific relationship and reach outcomes:
1. The Additive Framework (Value Expansion)
- Objective: Deepen the author's premise by contributing a practical mental model, an overlooked operational edge case, or concrete production metrics.
- Strategic Target: Building goodwill with respected peers and demonstrating technical depth to observers.
- Rhetorical Structure:
- Sentence 1 (Validation + Anchor): Acknowledge the core insight with a specific anchor ("Spot on regarding query latency...").
- Sentence 2-4 (The Additive Insight): Introduce an additional layer, production constraint, or real-world metric not mentioned in the post.
- Sentence 5 (Closing Synthesis): Summarize the practical takeaway for other practitioners reading the thread.
- What to Avoid: Merely restating or summarizing what the author already wrote in different words.
2. The Contrarian Framework (Respectful Challenge)
- Objective: Challenge an unexamined assumption, highlight a severe failure mode, or present the counter-narrative based on direct operational experience.
- Strategic Target: Establishing equal or senior status with the author; sparking lively debate that draws dozens of sub-replies.
- Rhetorical Structure:
- Sentence 1 (The Boundary Condition): Concede where the author's advice works, then introduce where it breaks ("This works brilliantly for Series B+ companies, but fails at early-stage...").
- Sentence 2-4 (The Empirical Evidence): Provide specific numbers, architectural friction, or dollar costs that disprove the universal claim.
- Sentence 5 (The Working Rule): Provide a balanced rule of thumb clarifying when to use Approach A vs Approach B.
- What to Avoid: Personal attacks, intellectual arrogance, dismissive sarcasm, or picking semantic fights.
3. The Question-Forward Framework (Engagement Anchor)
- Objective: Ask a piercing, diagnostic question that compels the post author and other senior practitioners to respond directly to your comment.
- Strategic Target: Maximizing the likelihood of an author reply, triggering algorithmic visibility to their 1st-degree connections.
- Rhetorical Structure:
- Sentence 1 (The Context Observation): Identify a specific operational tension or trade-off raised by the post.
- Sentence 2 (The Real-World Dilemma): Highlight the friction in executing the author's advice in production.
- Sentence 3 (The Diagnostic Inquiry): Ask how the author handles that exact friction under pressure.
- What to Avoid: Lazy questions ("Where can I learn more?", "Any tips for beginners?").
Algorithmic Dwell-Time & Quality Standards
LinkedIn's feed algorithm evaluates comments using dwell time, sub-thread velocity, and author responses:
- Word Count Sweet Spot (40–120 Words):
- Comments under 25 words rarely earn top-comment placement because readers scroll past them in under 1 second.
- Comments over 150 words get truncated behind a comment "...more" button and suffer high skip rates.
- The 40-120 word range guarantees full mobile visibility and maximum read completion.
- The First 60 Minutes Golden Rule:
- Comments left within the first 60 minutes of a post going live receive up to 5x more impressions than comments left at hour 12.
- Early, substantive comments frequently get pinned or upvoted to the permanent #1 comment position.
- Zero-Fluff Blacklist:
- Banned Openers: "Great post!", "Love this!", "Totally agree!", "100%", "Thanks for sharing, [Name]!", "Spot on!"
- Banned Sycophancy: "You are such an inspiration", "Another masterclass as always".
- Banned AI Vocabulary: Do not use delve, tapestry, leverage, game-changer, paradigm, or synergy.
How It Works Step-by-Step
When executing linkedin-comment-drafter, follow these sequential steps:
- Input Analysis & Parsing:
- Ingest the LinkedIn post URL or raw text content.
- If a URL is supplied, run
lib.url_parser.parse_post_url(url)to validate the URN and ensure it targets a validactivity,share, orugcPost. - Identify the author's primary thesis, industry context, audience seniority, and underlying tone.
- Objective & Persona Calibration:
- Determine whether the primary target is the post author, third-party readers browsing the comments, or potential sales prospects.
- Select the optimal style: Additive, Contrarian, or Question-forward.
- Multi-Option Drafting:
- Draft candidate comments representing all three frameworks.
- Verify that each option incorporates concrete terminology, avoids banned AI words, and maintains professional brevity.
- Approval Staging:
- Format the drafted options and stage the preferred option in
PubloraClientviaclient.comment(post_urn=..., text=..., approved=False). - Present the options to the user alongside the approval ticket token.
- Format the drafted options and stage the preferred option in
- Human Sign-off & Dispatch:
- Wait for explicit user confirmation. Upon confirmation, execute
client.approve_and_execute(token).
- Wait for explicit user confirmation. Upon confirmation, execute
Pre-Flight Comment Checklist
Before submitting a drafted comment to the approval gate, verify:
- Does this comment offer value to someone who never clicks my profile?
- Is it free of generic praise openers ("Great post", "Love this")?
- Does it avoid all banned words (delve, tapestry, leverage)?
- Is the length between 40 and 120 words?
- Does it contain at least one concrete anchor (a tool, metric, dollar amount, or framework)?
- Is the tone respectful, peer-to-peer, and devoid of unearned arrogance?
Approval Gate
No comment is ever published directly to LinkedIn without user confirmation:
[STAGED COMMENT APPROVAL]
Target Post URN: urn:li:activity:7123456789012345678
Target URL: https://www.linkedin.com/feed/update/urn:li:activity:7123456789012345678/
Selected Framework: Option B (Contrarian)
Word Count: 68 words
Drafted Text:
"This holds true when managing mature relational data models, but breaks down completely under event-driven streaming. When we decoupled our inventory ingestion from synchronous Postgres transactions to an append-only Kafka log, our deadlocks dropped to zero, but event consistency required an entirely new idempotency layer. How does your team manage transaction rollbacks across microservice boundaries without global locking?"
Approval Token: b294f83a-49c0-4382-b7e1-8899aabbccdd
Status: HELD IN APPROVAL GATE.
To dispatch to LinkedIn, reply: 'APPROVE b294f83a-49c0-4382-b7e1-8899aabbccdd'Advanced Comment Tactics: The Pin-Hunting Methodology
Top-tier creators frequently pin the single most valuable comment to the top of their post. Being pinned guarantees tens of thousands of impressions from visitors who never scroll down.
To optimize for pinning:
- The First-Sentence Anchor: Quote the author's exact phrase in quotation marks.
- The Hard Data Injection: Supply a specific dollar number, benchmark latency, or conversion percentage.
- The Micro-Case Study Structure:
- Problem faced in production.
- Counter-intuitive hypothesis tested.
- Metric delta achieved.
- Practical heuristic for readers.
- The No-Self-Promotion Rule: Never link your own tool or drop a sales pitch in the comment. Authority is earned through pure unselfish practitioner value.
Examples
Example 1: Additive Comment on an Engineering Leadership Post
Original Post Summary: Author discusses how daily 15-minute standups often degenerate into status updates and suggests replacing them with async Slack check-ins.
The transition to async standups only succeeds if you simultaneously remove the obligation to respond immediately.
When we switched to async written updates, engineers initially felt pressured to reply within 3 minutes of seeing a message, which was more disruptive to deep work than the original 15-minute Zoom call.
What made it stick was setting a strict "batch window": post your update by 10 AM, but nobody reviews or responds until 1:30 PM after morning focus blocks.Example 2: Contrarian Comment on a B2B SaaS Pricing Post
Original Post Summary: Influencer argues that every B2B startup should offer a free tier to build a bottom-up developer flywheel.
A free tier works wonders for developer utilities with zero marginal infrastructure cost, but it can quietly kill deep-tech and AI infrastructure companies.
When each user action consumes GPU compute or third-party API tokens, an unmonetized free tier creates a toxic illusion of product-market fit. You end up optimizing your architecture for hobbyists who will never have corporate purchasing cards.
We found that charging a non-trivial minimum ($250/mo) immediately filtered out tourists and attracted teams with real enterprise data scale.Example 3: Question-Forward Comment on a Hiring Post
Original Post Summary: VP of Talent explains why take-home coding challenges should replace whiteboard algorithm interviews.
Take-home projects reveal real-world engineering habits far better than whiteboard riddles, but senior candidates with families or demanding current roles increasingly decline to complete 6-hour unpaid projects.
How does your team keep candidate drop-off low among senior staff engineers? Do you scope the take-home to under 90 minutes, or do you compensate candidates for their review time?Example 4: Outbound Prospecting Comment on a Target CTO’s Post
Original Post Summary: Target prospect (CTO of a 200-person logistics startup) posts about dealing with rising cloud data egress costs across AWS regions.
Cross-region egress is one of the most opaque lines on an AWS bill. Most teams don't realize that NAT Gateway data processing fees often exceed the actual EC2 compute cost for microservices communicating across availability zones.
When we audited our traffic topologies, simply placing service replicas in the same AZ and routing via private DNS eliminated 40% of the monthly networking invoice in 48 hours.
Are your biggest egress spikes originating from inter-service API traffic or database replication?Example 5: Micro-Case Study on Developer Tool Adoption
Original Post Summary: Head of Engineering discusses why internal platform tooling rarely gets adopted by product teams.
"Internal developer portals fail when adoption is mandated rather than earned."
We made this mistake in 2024 by forcing 40 engineers to use a bespoke internal staging deployment CLI. Adoption stalled at 25% until we made one change: we instrumented the CLI to measure deployment cycle time and proved it shaved 9 minutes off every pull request verification.
Once developers saw empirical proof that the tool gave them 45 minutes of their afternoon back, voluntary adoption climbed to 96% in two weeks.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
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.

