Template · Agent · Official

Social posts written native to each platform

Turn published work into platform-native posts, schedule them, and collect the engagement numbers back into one report.

Get this template

Opens in AGNT and takes about a minute.

What this actually involves

Tools it needs	X (Twitter), YouTube, Notion, Google Sheets, Unsplash
Setup	4 steps, about ten minutes
Approval	Yours, per category — nothing is sent on your behalf unless you say so
Where it runs	Your own machine. Credentials stay in a local vault.

What the agent actually does

Every platform gets a post built for it

A long piece becomes a thread that stands on its own, a short post with a genuine hook, and a newsletter section — each written for how people read on that platform rather than the same paragraph reposted.

Schedule with approval

Drafts queue for a human to approve, then publish on the schedule you set. The approval step costs a minute and removes the entire class of problem where an unattended system posts something regrettable.

Close the loop

Engagement comes back into the same report as everything else, so what to write more of is a question with an answer instead of a matter of opinion.

The tools this job needs

X (Twitter)

Where the work arrives. The feed where your market thinks out loud. The agent watches it and reads what turns up in full, rather than matching a rule against a subject line.

YouTube

Context the agent pulls in before deciding. Long-form content worth repurposing everywhere else. It is read, not just referenced — which is what lets the decision account for it.

Notion

Context the agent pulls in before deciding. The team wiki agents can actually keep current. It is read, not just referenced — which is what lets the decision account for it.

Google Sheets

Where the result lands. The database every team already has open. Nothing is written here until the agent has formed a view and, where you asked for it, you have approved it.

Set it up

Connect the accounts you publish to and the source your long-form work lives in.
Give the agent the voice skill so adaptations sound like you on every platform.
Route drafts to an approval queue with the schedule attached.
Pull engagement back on a cadence and read it alongside the rest of your reporting.

You can see exactly what it did

Nothing happens behind your back.

Every run leaves a receipt: what the agent read, which tools it called, what it decided and why, and precisely what it changed. Anything irreversible — sending, paying, publishing, deleting — waits for you to approve it. It runs on your own machine, with your own credentials, and the whole trail is yours to read afterwards. The point is not that you trust it. The point is that you never have to.

The brief it works from

This is the actual instruction set the template installs — what the agent is told to do, and what it is told never to do. Every line of it is yours to edit in AGNT after install.

Read the full brief

You turn work the user has already published into platform-native posts, and you report what happened to them.

Adapt, never cross-post

Each platform is a different medium. A thread needs a first line that earns the second and one idea per post. A short post needs the whole thought to land at once. A professional network wants a different register entirely. Rewrite for each — the same paragraph pasted four times is why so much social content underperforms, and it is visible to every reader.

Write in the user's voice

Load their style guide. Match how they actually write, including what they never say. Do not add hashtags unless they use hashtags. Do not add emoji unless they use emoji. Do not open with a rhetorical question, and never use the engagement-bait phrasings — "let that sink in", "here's the thing", "unpopular opinion" — unless that is demonstrably how they write.

Everything queues for approval

You draft and schedule; a human releases. The minute that costs is cheap insurance on anything public, and it removes the entire class of problem where an unattended system posts something regrettable at 3am.

Never

Never reply to anyone, never engage in an argument, never DM, never follow or unfollow, and never post about anything sensitive, political or newsworthy without a human explicitly deciding to. If a draft touches a live news event, flag it rather than queueing it — context changes hourly and you will not know that it has.

Close the loop

Pull engagement back on a schedule and report it alongside the rest of the user's numbers. Say which formats and topics performed, with the actual figures. Without this the content decision is made on instinct forever, and the whole exercise stays an activity rather than something anyone can steer.

Shown exactly as it ships. The agent inherits whichever model you already use.

Cross-posting is why so much social content underperforms

The same text pushed to four platforms is optimised for none of them. A thread needs a first line that earns the second; a short post needs the whole idea to land at once; a professional network rewards a different register entirely. Adapting properly is genuinely a rewrite each time, which is exactly why it stops happening in a busy week — and exactly the kind of work worth handing to an agent.

Consistency beats intensity, and consistency is an ops problem

Presence compounds when it is regular, and regularity fails for operational reasons rather than creative ones: the person who does it is busy, or on holiday, or shipping. A pipeline that reliably produces drafts from work you were already doing turns a discipline problem into an approval queue, which is a much easier thing to sustain.

Reporting closes the loop or the loop stays open

Without engagement flowing back, the content decision is made on instinct forever. Pulling the numbers into the same report as everything else is unglamorous and it is what turns publishing from an activity into something you can actually steer.