Linkedin MAXXING With Gemini CLI

If you give AI a topic and ask it to write a LinkedIn post that leaves out the work that makes the post worth reading: understanding your buyers, finding an interesting angle, pulling together evidence, and knowing what you can credibly say.

These 12 prompts turn that work into a repeatable system, from researching your market to building a content calendar and identifying which posts contribute to real business conversations.

Here's what you'll build:

• A voice guide based on how you actually communicate.
• A research process that finds useful ideas before you need them.
• A draft and review workflow with specific quality standards.
• A reporting loop that helps you decide what to write next.

Save this for your next content planning session.

Before You Run a Single Prompt: Load Your Business Context

Gemini CLI needs more than your job title and a request to "sound human." Give it enough context to understand who you sell to, what you believe, and which claims you can support.

Paste this first:

Use the following as the working context for my LinkedIn content. Save it in a reference document I can reuse, and tell me where it is stored. Business: [Company name, website, product or service, pricing, and typical sales cycle]. Audience: [Buyer roles, company sizes, industries, and locations]. Their main problems are [problems], their common objections are [objections], and the events that make them look for help are [triggers]. Positioning: We help [audience] achieve [outcome] through [approach]. Our strongest differences are [differences]. We are not a good fit for [disqualifiers]. Point of view: I believe [three opinions about my industry]. I disagree with [common advice] because [reason]. Evidence: Use these approved case studies, results, examples, and personal experiences: [sources]. Do not invent customers, revenue figures, stories, or quotations. Voice: Here are [10-20 posts] and [transcripts or writing samples] that sound like me. Avoid [phrases, tones, and formatting habits]. Goals: My primary LinkedIn objective is [qualified conversations / deal support / recruiting / awareness]. My offer is [offer], and the next step I want interested readers to take is [action]. Working rules: Use connected sources only where access is available and authorized. If a source is inaccessible, ask for an export or pasted material. Keep LinkedIn publishing, comments, connection requests, and messages with me. Draft into [review destination], distinguish evidence from inference, and flag anything that needs my input.

Keep that context current as your offer, audience, and opinions change.

PART 1: RESEARCH + CONTENT PRODUCTION

Prompts 1-8

1. Build Your Voice Guide

A useful voice guide should explain the choices behind your writing, including how you introduce an argument, use examples, and move between ideas.

"Conversational and authoritative" won't get you very far.

Analyze my supplied posts and transcripts to build a practical LinkedIn voice guide. Identify: How I open posts, explain ideas, and close an argument. My usual vocabulary, sentence lengths, paragraph lengths, and use of lists. How I express disagreement, uncertainty, humor, and confidence. Phrases I use naturally, with verbatim examples and source references. Writing habits that appear in weaker examples but should not become rules. Create a list of phrases and patterns to avoid, including unnecessary repetition, manufactured drama, unsupported certainty, and excessive one-line paragraphs. Finish with three before-and-after examples showing how you would edit generic copy into my voice without changing its meaning. Keep the guide specific enough that another writer could apply it.

Why this matters: A good voice guide makes feedback reusable. Instead of correcting the same phrasing every week, you can improve the instructions that shape every draft.

2. Find the Buyer Questions Worth Answering

Your sales calls and customer conversations can reveal questions that never show up in a keyword tool. The useful ones usually involve a decision someone is struggling to make.

Review the supplied sales calls, support conversations, customer interviews, and closed-lost notes from [period]. Group recurring questions into: Problems buyers are trying to understand. Concerns that delay a purchase. Alternatives they are comparing. Expectations that lead to disappointment. Questions they ask immediately before buying. For each theme, include an anonymized quotation, source reference, number of distinct conversations where it appears, buyer role, and stage in the buying process. Suggest two LinkedIn angles per theme: one that helps the reader diagnose the problem and one that helps them evaluate a solution. Rank the top 10 by recurrence, commercial relevance, and the strength of our available evidence. Separate explicit buyer statements from your interpretation.

Why this matters: This gives you a content backlog tied to conversations your business is already having, with language you can recognize and evidence you can check.

3. Find Topics Moving From X Into Your Market

A discussion gaining attention on X can become a useful LinkedIn post, provided you explain why it matters to your buyers and contribute something beyond a rewrite.

Using accessible public sources and any search tools already connected, review the last [seven days] of discussion around [topics] and these accounts: [accounts]. Find up to 10 developments or arguments relevant to [buyer]. For each, return: The original source, author, publication date, and link. A neutral summary of what happened or what is being argued. The strongest competing interpretations. The practical implication for our audience. A possible angle supported by our experience or evidence. Label engagement figures as a snapshot, avoid treating popularity as proof, and check factual claims against original sources where possible. Recommend three ideas worth developing, explaining what I would need to contribute before each becomes a credible LinkedIn post.

Why this matters: Research gives you a starting point; your judgment gives readers a reason to follow you. A useful adaptation connects the discussion to a decision they actually face.

4. Analyze Posts That Outperform Their Authors' Usual Results

Comparing every post with a celebrity account's biggest hit will distort your research. Compare a post with that author's typical results first, then examine what might explain the difference.

Analyze this supplied LinkedIn swipe file and the available performance data: [file]. Where enough comparable data exists, compare each post's engagement with the author's median for a similar format and period. Show the calculation and flag small samples, missing metrics, and unusually recent posts. For the strongest outliers, identify: The topic and intended reader. The opening, argument, supporting evidence, and closing. The format and approximate length. Whether the discussion suggests genuine buyer relevance. Possible confounding factors, including a launch, news event, or unusually large distribution. Produce five reusable structural patterns, then propose an original idea for each using my own expertise. Do not copy distinctive wording or present an observed pattern as an algorithm rule.

Why this matters: You get a more useful set of hypotheses to test, while avoiding the assumption that copying a successful post's layout will reproduce its results.

5. Turn Recorded Expertise Into Original Posts

A transcript can contain a strong post buried inside a story, an offhand explanation, or a disagreement that deserves more attention than it received in the original conversation.

Review this interview, podcast, webinar, or meeting transcript: [source]. Find eight moments that could support an original LinkedIn post. Prioritize decisions, mistakes, surprising observations, practical explanations, and opinions backed by experience. For each moment, provide: The exact quotation and timestamp or source location. The main idea and who would benefit from it. Any context needed to avoid misrepresenting the speaker. Supporting facts already available. One follow-up question that would make the idea stronger. Select the three strongest moments and draft a post from each using my voice guide. Preserve the speaker's level of certainty, remove confidential details, and leave placeholders wherever a result or example needs confirmation.

Why this matters: You can produce content from expertise that already exists without asking someone to manufacture a fresh opinion every time the calendar needs filling.

6. Build a Four-Week Content Calendar

A calendar should balance what buyers need to understand with what you have enough evidence to say. Filling every slot with a different version of your offer makes the feed repetitive.

Build a four-week LinkedIn calendar at [three posts per week] using the approved ideas from our research. Include a mix of: Practical explanations that help buyers solve a problem. Opinions supported by experience or evidence. Customer examples we have permission to share. Answers to objections and buying questions. Occasional offer posts with a relevant next step. For each post, specify the reader, single objective, central argument, source material, suggested format, and intended reader action. Check for repeated topics, overlapping arguments, unsupported claims, and too many promotional posts. Write full briefs for week one and shorter outlines for weeks two through four. Leave room for one timely replacement each week rather than assuming the calendar must remain fixed.

Why this matters: A useful calendar reduces weekly planning work while leaving room for better ideas, new evidence, and changes in the market.

7. Draft Posts With Something to Say

The research should determine the argument before the bot starts polishing sentences. Otherwise, you can end up with clean writing that takes several paragraphs to make an obvious point.

Write three LinkedIn drafts from these approved briefs: [briefs]. Use my business context and voice guide. For each draft: Make one clear argument for one intended reader. Open with a specific observation, tension, result, or question supported by the brief. Explain the reasoning with a concrete example. Use bullets where they make parallel points easier to scan. Vary sentence and paragraph length, allowing occasional one-line emphasis. End with a useful implication or natural next step. Aim for [180-300 words] unless the idea warrants a different length. Avoid repeated conclusions, invented personal stories, exaggerated promises, and formulaic contrasts. Supply two alternative openings and explain briefly how they change the emphasis. Flag missing evidence separately from the publishable copy.

Why this matters: Clear constraints give you drafts that are easier to judge. Your review can focus on the quality of the argument instead of repairing unnecessary filler.

8. Create a Separate Editorial Review

Give the reviewer a defined job: find weaknesses in the draft and explain how to fix them. A reassuring "This is strong" is not an editorial process.

Review these LinkedIn drafts against the brief, source material, business context, and voice guide. Score each out of 100 using: Buyer relevance: 25 points. Specificity and supporting evidence: 25 points. Strength and clarity of the argument: 20 points. Match with my voice: 20 points. Readability and formatting: 10 points. Show the evidence behind each score. Check every factual claim, identify generic passages, and highlight repetitions, abrupt one-line fragments, and paragraphs that would work better as lists. Return a publishable revision, a short explanation of meaningful changes, and any unresolved questions. Treat invented evidence, exposed confidential information, or a misrepresented source as a blocking issue regardless of score. Keep everything in the review document for human approval.

Why this matters: A review rubric makes your standards explicit. The score helps organize feedback, while the source checks and editor's judgment determine whether the post is ready.

The first hour after you publish

The review in prompt 8 ends when you have a publishable revision and you approve it. The next step is the post going live. The first hour after you publish is when LinkedIn is still deciding who else sees it. Reactions in that hour, from people who actually read the post, are what carry it past your own connections.

Do these three things as soon as the post is live.

Ask everyone you can reach. Friends, people you know, colleagues, clients, and anyone else you can actually message. Ask them to like the post and to interact with it: a reaction, a comment that says something, or a repost if they mean it. One personal message when it is live. A personal ask works. A mass message does not.

Use engagement groups. Members engage with each other's posts. The groups in the next section are the same list already published on our LinkedIn engagement groups page, copied here so you do not have to go looking. Drop your link, do your part of the exchange, and leave the chat. Check the pinned message before the first post. Each group is run by its own admins.

Use upvote.club for the first engagement. Upvote.club is the best platform for this first hour. You get real engagement on the post you just published, without staying in a group chat. Start free on upvote.club.

Gemini CLI can run the upvote.club part. Upvote.club has an MCP server at https://mcp.upvote.club/mcp. Connect it once. After you approve the post, Gemini CLI can create the engagement task from the post URL. You still publish, comment, send connection requests, and send messages yourself. The bot places the task.

In Gemini CLI, add the server with this command:

gemini mcp add upvote-club https://mcp.upvote.club/mcp --transport http

LinkedIn engagement groups you can join

Free communities we checked on Oct 21, 2026. Member counts are what each platform showed that day. Each group is run by its own admins, so check the pinned message before your first post. Want real engagement on your own posts without searching through groups? Start free on upvote.club.

PART 2: PIPELINE + PERFORMANCE

Prompts 9-12

9. Build Content Around Stalled Deals

Some of your most useful content may answer a concern that several prospects have raised privately. The goal is to explain the issue well enough that future buyers arrive better informed.

Review the supplied CRM records and call notes for qualified deals that stalled during [period]. Identify recurring objections and distinguish: Explicit objections recorded in conversations. Possible explanations that remain unconfirmed. Operational delays that content is unlikely to solve. For each repeated objection, record the buyer role, deal stage, available evidence, and the question a useful LinkedIn post should answer. Recommend five content briefs. Each should explain the concern fairly, show where our approach is appropriate, acknowledge where it is not, and identify the proof required. Draft the two strongest posts without naming prospects or revealing deal details. Separately, suggest which approved existing resources a salesperson could share in a relevant conversation.

Why this matters: This connects marketing with the questions salespeople repeatedly need to answer. It also helps you avoid writing persuasive content about problems your buyers do not consider important.

10. Prepare Relevant Conversation Briefs

When someone responds to a post, the best next step depends on what they said and whether a business conversation would be welcome. Engagement alone does not establish buying intent.

Review these LinkedIn conversations I have supplied: [comments, inbound messages, or manually recorded interactions]. Group them into: Questions that deserve a useful public answer. Conversations where sharing a resource would help. Explicit interest in our offer. General discussion with no commercial next step. For each relevant conversation, summarize the person's stated need, available business context, and what remains unknown. Draft a response that addresses their actual message before introducing anything else. Where appropriate, suggest one follow-up question or a resource we already have. Do not infer purchase intent from a like, invent familiarity, or force a meeting request into a conversation that does not support one. Keep all replies as drafts for me to review and send.

Why this matters: Preparation can make your responses more thoughtful without turning every interaction into a sales sequence. Sometimes the right outcome is simply a useful answer.

11. Report on Business Results With Clear Attribution

Impressions help you understand distribution, but they cannot tell you whether the right people became interested or whether a post contributed to a deal.

Combine my supplied LinkedIn analytics, tracked website visits, inbound inquiry records, and relevant CRM data for [period]. Report on: Reach and engagement by post, topic, and format. Qualified inbound conversations. Resource sign-ups and demo requests where tracking exists. Opportunities with documented LinkedIn touchpoints. Content production time and review effort. Separate directly tracked conversions, self-reported discovery, and possible influence. Do not attribute an entire deal to LinkedIn because one contact engaged with a post. Highlight posts that attracted relevant buyers, posts that reached a broad but less relevant audience, and posts with insufficient data to judge. Finish with three findings, three limitations, and one recommended experiment. If possible, assess posts at consistent ages so newly published content is not unfairly compared with older posts.

Why this matters: Better reporting gives you a reason to repeat or change an approach. It also makes uncertainty visible instead of turning incomplete tracking into a confident revenue claim.

12. Turn Feedback Into a Better System

Every edited draft contains information about your standards. Capture the recurring lessons so you spend less time correcting familiar problems.

Compare the original drafts, my edits, final published versions, and available performance data from [period]. Identify: Voice corrections I have made repeatedly. Claims that needed more evidence. Ideas that became clearer after a change in framing. Formatting changes that improved readability. Topics that generated useful buyer conversations. Propose updates to the voice guide, research briefs, and editorial checklist. For every proposed rule, show an example, explain its purpose, and note when an exception may be appropriate. Keep stylistic preferences separate from performance hypotheses. Do not create a permanent rule from one successful or unsuccessful post. Return a short change log for my approval and suggest one controlled experiment for the next month.

Why this matters: This is how the workflow becomes easier to manage over time. Your corrections become usable guidance, while performance ideas remain testable rather than hardening into superstition.

How to Use All 12 Prompts

Build the workflow in stages so you can inspect the output before adding another recurring task.

Week 1: Establish the foundation. Load your business context, run prompt 1, and check the voice guide against writing you would actually publish.

Week 2: Build your research backlog. Run prompts 2-4 using a manageable set of customer conversations, market sources, and supplied LinkedIn examples.

Week 3: Produce the first batch. Use prompts 5-7 to turn approved material into a calendar and a small number of drafts.

Week 4: Strengthen the review process. Run prompt 8, publish the posts you approve, run the first hour on each post you publish, and record the changes you make.

Weeks 5-6: Connect content with conversations. Add prompts 9 and 10 where you have enough sales context and inbound activity to make them useful.

Then repeat monthly: Use prompts 11 and 12 to review results, improve the instructions, and choose the next experiment.

For any recurring job, define its inputs, destination, frequency, and owner. Test a run before relying on the schedule, and have the bot report missing access or incomplete work.

You do not need 12 separate bots on day one.

Start with the job that consumes the most time and produces an output you can reliably evaluate.

The Real Talk

Gemini CLI can help you organize research, retrieve useful evidence, prepare drafts, and learn from your edits. The quality still depends on the experience and judgment you bring into the system.

That means having a real opinion, being willing to explain it, and noticing when a polished draft says something you would never defend in a conversation.

If you want the first engagement on the posts this workflow produces, start free on upvote.club.

Otherwise, load your context and run prompt 1. Improving the inputs will make every step after it more useful.

Read next

• LinkedIn engagement pods
• Linkedin MAXXING With Claude
• Linkedin MAXXING With Cursor
• Linkedin MAXXING With ChatGPT
• Linkedin MAXXING With ChatGPT Dots
• Linkedin MAXXING With Claude Code
• Linkedin MAXXING With Codex
• Linkedin MAXXING With Grok
• Linkedin MAXXING With Grok Bot
• Linkedin MAXXING With Muse
• Linkedin MAXXING With OpenCode