Loop · Content
Run LinkedIn content in Maskin so your team keeps its voice and the final say while Maskin reads what happened this week, plans the calendar, drafts every post, and keeps a steady posting rhythm going.
The job
Keep a founder-voice LinkedIn presence going from the moments already happening inside your business: a shipped feature, a customer win, a competitor move, a hard lesson. The result is a steady posting rhythm, planned two weeks at a time, and founder time back for the work only founders can do, instead of writing posts cold at 11pm on Sunday. The work is done when the post is live on LinkedIn and the next retro has read how it did, not when a draft is generated.
Replaces: ghostwriter retainers · standalone “AI LinkedIn post generator” tools · a shared doc of half-drafted posts that decays · asking the founder to remember what to post about
Stays human: the stories and opinions only you hold · the voice · the approval on every post · which drafts get killed
Human ↔ Maskin split
| Your team keeps | Maskin runs |
|---|---|
| Answering the interview questions with the real stories and opinions | Reading the last two weeks of shipped work, feedback, and signals, then writing the questions |
| The voice each post is written in | Planning the calendar and drafting the whole fortnight in one pass, in the right voice per account |
| The approval on every post | An editor agent checking every draft before it reaches you |
| The judgement to kill a weak draft | Reworking the drafts that get sent back |
| On some accounts, pasting the approved post into LinkedIn | Posting for you on the others, then a retro every two weeks that reads how posts did |
How the loop runs
- Every two weeks, Maskin reads the last fortnight: what shipped, what feedback came in, and how recent posts did.
- From that, it writes interview questions for each account owner, so every post starts from something true.
- The account owners answer. The real stories, opinions, and numbers only they hold are the raw material, and this is the human step the whole fortnight hangs on.
- Once the answers are in, Maskin plans a calendar of up to 10 posts across personal accounts and the company page, and drafts all of them in one pass, each in that account’s voice, with media noted where a visual helps.
- An editor agent reads every draft the way a busy follower would and sends weak ones back for rework. Then the account owner reads each post, tweaks or kills it, and approves it.
- Approved posts go live on their scheduled day. On some accounts Maskin posts for you, and on others the owner pastes the approved text into LinkedIn.
- Two weeks later, the retro reads how the posts did and files what to do more and less of, and the next plan starts from it.
Trust surface
- Voice is a workspace skill, not a prompt. Each account (a founder’s personal profile, a teammate’s, the company page) has a voice skill the loop applies. If an account has no voice skill yet, one is drafted and approved by its owner first. Nobody posts in a voice its owner hasn’t signed off.
- Nothing goes out unapproved. An editor agent checks every draft, and the account owner approves each post before it goes live.
- Weak drafts get killed, not shipped. The editor sends weak drafts back for rework, and a draft that can’t be saved is dropped instead of posted.
Run it
Book a demo: see the loop running on the Maskin Team’s own LinkedIn profiles and company page, then decide whether to run it on yours.
What’s inside the loop
Loop
A closed cycle: something fires it (a signal, a schedule, a click), agents do the work, a human gates the taste-sensitive parts, and it ends on a concrete outcome, not on done for now. Every listing on the Marketplace is a loop.
Maskin loops carry typed outcomes and verdicts that persist across tools, not recurring workflow execution inside one product.
Agent
A named role a model performs: an SEO analyst, a discovery synthesizer, an outbound writer. Not an AI teammate: a role with a job description, tools it can use, and gates it has to pass before its work leaves the workspace.
MCP
The wire an agent uses to reach the tools your team already uses: the CRM, the docs, the inbox, the analytics. If a human on your team can log in and click, an MCP-native agent can be given the same access under the same gates.
Skill
A packaged capability you attach to an agent: draft a follow-up in our voice, summarize a customer call against our discovery template. Skills are the reusable pieces of judgment your team is already building without realising it.
Inside a loop listing, agents / MCPs / skills appear as “built with” components of the loop, never as top-level marketplace sections of their own.
FAQ
Can I use AI to create content for LinkedIn?
Yes, as long as the substance is yours. In this loop the account owner answers interview questions about what actually happened, a voice skill for that account shapes the draft, an editor agent checks it, and the owner approves every post before it goes live. The AI does the reading, planning, and drafting. You bring the stories and keep the final say, including the right to kill a draft.
Why is there so much AI-generated content on LinkedIn?
A post written from a prompt alone has nothing real behind it, so it reads like every other one. This loop starts from the other end: what your team shipped, what feedback came in, and what the account owners say in their interview. An editor agent checks every draft, and the owner can drop any draft that falls short before it goes live.