Content Repurposing Agent
One asset into every channel, without inventing a single claim
- 8hard rules
- Things it must never do, enforced in every run
- 4handover rules
- Conditions where it stops and gives the work to a person
- 5output fields
- A structured result, not a paragraph you have to re-read
- ≤4steps per run
- Bounded work: at most 2 tool calls, enforced by the runtime
Read-only by design. This agent has no path to sending, spending, publishing or committing anything. Where that limit is the product, removing it would remove the reason to trust the output.
Overview
Turning one article into ten posts is the most-copied marketing automation there is, and the copies share a flaw. Compressing a thousand words into a hundred means choosing what to drop, and a model that drops a qualifier keeps the claim underneath it. "In our tests, most users saw faster loads" becomes "2x faster". Nobody wrote that, and it is now published under your name.
This agent may only subtract. Every line it produces traces to a sentence in the source, and it reports the qualifiers and caveats that would not fit — because those omissions are a decision someone should make deliberately rather than discover later.
It writes for the channel rather than reformatting: what works as a thread does not work as a newsletter opener, and the same paragraph cut three ways is not repurposing. It drafts; it publishes nothing.
Benefits
Ten channels from one effort
The work you already did, everywhere it belongs.
Nothing you have to retract
It cannot publish a claim you never made.
Consistent everywhere
The same facts, in each channel's own form.
What it does
Can only subtract
Every line traces to a sentence in the source. It cannot introduce a fact, a figure, or a claim.
Reports what it dropped
The qualifiers that would not fit are listed, so leaving them out is a decision, not an accident.
Writes for the channel
A thread, a newsletter opener and a caption are different forms, not the same text re-cut.
Keeps the claim with its evidence
Where a claim cannot fit alongside its basis, it drops the claim rather than the basis.
Matches your voice
Give it past posts and it follows how you actually write.
Publishes nothing
Drafts for review. It has no access to any channel.
What this changes
What this changes about your week
- It takes over
- Cutting one asset into versions for every channel.
- Done today by
- A marketer, for an afternoon per asset.
- How often
- Every piece you publish.
- Typical manual effort
- One article into six channel-native versions is half a day.
Why Content in particular. It can only subtract, so it cannot publish a claim you never made — which is exactly what happens when a generator drops the qualifier and keeps the number.
Runs unattended
Started by you or by an event, and it finishes on its own. Nothing waits for someone to be at a desk.
The same standard every time
The two-hundredth item is held to the bar the first one was. Consistency is the part people cannot sustain.
It cannot act on its own
Content Repurposing Agent has no path to sending, spending or committing. That limit is why its output is safe to act on.
How it runs
This agent runs server-side through the PROMIVO runtime. Each run is logged step by step and every tool call is permission-checked before it executes.
- 1InputValidated against the agent input schema
- 2Modelclaude-sonnet-5 with automatic fallback
- 3Toolsextract_structured_data
- 4PermissionsRuns READ actions automatically; anything higher waits for approval
- 5OutputStructured result matching the output schema
- 6LoggingTokens, duration and every step recorded
Demo
Demo dataIllustrative sample output, abridged.
Input
{
"source": "…benchmark article…",
"channels": [
"linkedin",
"newsletter"
]
}Output
{
"outputs": [
{
"channel": "linkedin",
"content": "We ran the migration on 40 customer workloads. Median build time fell from 94s to 61s — on the projects that were already using the new compiler. The rest saw no change.",
"tracesTo": [
"Paragraph 4: median 94s → 61s",
"Paragraph 6: 11 projects unchanged"
],
"droppedQualifiers": [
"The 'no change' group was 11 of 40 — the post says 'the rest' rather than the count."
]
}
],
"escalate": false,
"disclaimer": "Derived only from the source. No fact, figure, date or claim has been added, no qualifier has been dropped to strengthen a claim, no number has been restated, and what had to be left out is listed.",
"notProduced": [
{
"why": "The source has no narrative or example to build a script around.",
"channel": "video-script"
}
]
}Requirements
Integrations
No integrations required.
Before you start
- The source content
- The channels you want it for
- Requires the Starter plan or higher
Use cases
After publishing
One article into social, newsletter and video script.
A webinar or a talk
Turn the transcript into everything else.
A case study
Cut it for each channel without overstating the result.
Pricing
Starter
Billed monthly through your PROMIVO subscription. Cancel at any time.
Runs consume your plan allowance for agent executions and tokens. See plan limits
FAQ
Why can it not add anything?
Because that is where repurposing goes wrong. Added context sounds helpful and is invented, and a claim published under your name is your claim whoever wrote it.
What if the source is thin?
It says so and produces less. Padding a thin source is how a post ends up saying something the source does not.
Can it post for me?
No. It drafts. Publishing is a decision, and a wrong post is not retractable once it is out.
Reviews
Reviews open once customers have run this agent.
Need this agent to work differently?
Tell us what to change and our team will scope a customised version for your business.
Customize this agent