Caelan's Domain

Agent: Content Repurposer

Created: April 16, 2026 | Modified: April 16, 2026

Cowork Features
Used: Agents, Skills, Rules

This is Part 9 of a 16-part series on building your AI VP of Marketing with Claude Cowork. Previous: Agent: Campaign Strategist | Next: Wiring It Together: Your First Marketing Pipeline

Quick Start
This article builds on the Campaign Strategist agent from Article 8. If jumping in here, you need a Cowork project with CLAUDE.md (Article 1), brand voice rules (Article 4), and at least one piece of voice-checked content to repurpose.

Starter config
A completed CLAUDE.md with your business context, a .claude/rules/brand-voice.md file with your voice guidelines, and a published blog post or article to use as source content.

One Piece, Many Formats

You know how to build Agents. This one multiplies your content.

Most small businesses treat content as single-use. You write a blog post, publish it, share it once on social media, and move on. That blog post took two hours to research, draft, and polish. It lives on your website, gets some search traffic, and that is the end of its useful life.

That is leaving value on the table. A single blog post contains enough material for an email to your subscriber list, a set of ad headlines for paid campaigns, a landing page section for a dedicated offer, and a newsletter blurb that drives traffic back to the full piece. Four additional touchpoints from work you already did. You are not creating more content -- you are extracting more value from content that already exists.

The catch is that each format has different rules. An email body needs a subject line and a single call to action. Ad headlines need to land in under 90 characters. A landing page section needs a hero headline that works in isolation. A newsletter blurb needs to hook readers in two sentences. Doing this by hand for every piece of content is tedious enough that most people skip it entirely.

Here is where a Content Repurposer agent earns its keep. You feed it a voice-checked, approved blog post -- one that already sounds like your brand -- and it produces all four derivative formats in one pass. Each output follows the constraints for its format. Each one maintains your voice. You review the batch, make corrections, and publish across channels.

Take the Tideway Bookkeeping example from earlier articles. Tideway publishes a blog post called "5 Tax Mistakes Freelancers Make." That post becomes an email sequence educating subscribers about each mistake, ad headlines targeting freelancers during tax season, a landing page section for Tideway's tax-prep service, and a newsletter teaser that pulls readers into the full article. One afternoon of writing, four channels of distribution.


Define Output Formats

Before you build anything, get specific about what the agent produces. Vague output requirements create vague output. Each format has its own structure, length constraints, and purpose -- and the reasoning behind those constraints matters more than the numbers themselves.

Email copy. The subject line does most of the work -- if no one opens the email, the body does not matter. Preview text is the second line of defense, the snippet that shows up next to the subject line in most inboxes. A single CTA keeps the reader focused. Two CTAs split attention and reduce clicks on both.

Ad headlines. You are paying per impression, so every word earns its place. The variants exist because you will test them against each other -- one headline might outperform the others by a wide margin, and you cannot predict which one without running them.

Landing page section. This is not a full landing page -- it is a section you drop into a campaign page. The headline needs to work without context, because visitors scanning the page will read it before anything else.

Newsletter blurb. It lives inside a weekly or monthly roundup alongside other content. Its job is to make the reader click through to the full article. It is a trailer, not a summary -- it should create enough curiosity to earn the click without giving away everything.

Same source material, four different jobs. Each format serves a different moment -- inbox, scroll, click, reminder -- and each one needs content shaped for that moment.


Build with /skill-creator

Open your Cowork project and type /skill-creator. This is the fastest path to a working agent. You give it a description of what you want, and it builds the agent configuration for you.

Here is the prompt to paste into /skill-creator:

Create an agent called "Content Repurposer" with the following specifications:

ROLE: Takes a voice-checked, approved piece of content (blog post, article, or guide) and transforms it into multiple marketing formats while maintaining brand voice.

INPUT: The full text of an approved blog post or article. The content has already passed brand voice review.

OUTPUTS (produce all four for every input):

1. EMAIL COPY
   - Subject line (6-10 words, no clickbait punctuation)
   - Preview text (40-90 characters)
   - Body (200-300 words)
   - Single CTA with specific action and benefit
   - Tone: conversational, direct, like writing to one person

2. AD HEADLINES
   - 5 variants, each under 90 characters
   - Mix of benefit-led and curiosity-led approaches
   - No exclamation marks, no ALL CAPS
   - Suitable for LinkedIn, Facebook, and Google Ads

3. LANDING PAGE SECTION
   - Hero headline (under 10 words)
   - Supporting paragraph (2-3 sentences expanding the headline)
   - CTA button text (2-5 words)
   - Must work as a standalone section on a campaign page

4. NEWSLETTER BLURB
   - 2-3 sentences that create curiosity without summarizing everything
   - CTA link text
   - Designed for inclusion in a weekly/monthly email roundup

RULES:
- Follow all brand voice guidelines in .claude/rules/brand-voice.md
- Follow content standards in .claude/rules/content-standards.md
- Maintain the same brand personality across all four formats
- Adapt tone for each format (email is warmer, ads are punchier, landing page is direct, newsletter is casual) while staying within brand voice boundaries
- Never invent claims, statistics, or benefits not present in the source content
- Present all four outputs in a single response, clearly labeled
- After each output, include a brief note on what to verify before publishing

Cowork builds the agent. What you get back is a configuration that looks something like this:

# Content Repurposer

## Role
Transform approved content into email copy, ad headlines, landing page
sections, and newsletter blurbs while maintaining brand voice consistency.

## Instructions
You receive voice-checked content that has already been approved. Your job
is to extract the core message and repackage it for four distinct marketing
formats. Each format has specific structural requirements and character
limits listed below.

Read .claude/rules/brand-voice.md before generating any output. Every
derivative piece must pass the same voice standards as the source content.

### Output Format: Email
- Subject line: 6-10 words
- Preview text: 40-90 characters
- Body: 200-300 words
- One CTA only
- Conversational tone, second person

### Output Format: Ad Headlines
- 5 variants under 90 characters each
- No exclamation marks or ALL CAPS
- Mix benefit-led and curiosity-led

### Output Format: Landing Page Section
- Hero headline: under 10 words
- Supporting paragraph: 2-3 sentences
- CTA button text: 2-5 words

### Output Format: Newsletter Blurb
- 2-3 sentences plus CTA link text
- Create curiosity, do not summarize

### Constraints
- Never add claims or data not in the source
- Flag anything that needs verification
- Present all four outputs in one response

The specifics of your generated configuration may vary. What matters is that the four output formats are defined with their constraints, the brand voice rules are referenced, and the agent knows not to invent content that is not in the source material.

Under the hood: Manual build steps
If you want to understand what /skill-creator automates, here is how to build the Content Repurposer agent by hand.

Step 1: Create the agent file. In your Cowork project, create a new file at .claude/agents/content-repurposer.md. This is where the agent's instructions live.

Step 2: Write the agent definition. Paste the following into the file:

# Content Repurposer

## Role
You are a Content Repurposer. You take approved, voice-checked content
and transform it into multiple marketing formats. You do not create new
content -- you repackage existing content for different channels.

## Context
Before generating any output, read:
- .claude/rules/brand-voice.md (voice guidelines)
- .claude/rules/content-standards.md (formatting standards)

## Input
You will receive the full text of an approved blog post, article, or
guide. This content has already passed brand voice review. Your job is
to preserve its message and voice while adapting it to four formats.

## Outputs
For every piece of input content, produce ALL of the following:

### 1. Email Copy
- **Subject line**: 6-10 words. No clickbait punctuation, no ALL CAPS.
- **Preview text**: 40-90 characters. Works alongside the subject line
  in inbox preview.
- **Body**: 200-300 words. Conversational, second person ("you").
  One core message. Lead with the value to the reader.
- **CTA**: One call to action. Specific action + specific benefit.
  "Read the full guide" is acceptable. "Learn more" is not.

### 2. Ad Headlines
- 5 variants, each under 90 characters
- Mix approaches: 2-3 benefit-led ("Cut your tax prep time in half"),
  2-3 curiosity-led ("The tax mistake 80% of freelancers make")
- No exclamation marks. No ALL CAPS. No clickbait.
- Suitable for LinkedIn Ads, Facebook Ads, Google Ads

### 3. Landing Page Section
- **Hero headline**: Under 10 words. Must work without surrounding context.
- **Supporting paragraph**: 2-3 sentences expanding the headline. Specific
  benefits, not vague promises.
- **CTA button text**: 2-5 words. Action-oriented.

### 4. Newsletter Blurb
- 2-3 sentences that create curiosity. Do not summarize the full piece --
  give enough to earn the click.
- **CTA link text**: Short phrase for the hyperlink.

## Rules
- Follow all guidelines in .claude/rules/brand-voice.md without exception
- Never invent statistics, claims, or benefits not in the source content
- Maintain consistent brand personality across all four formats
- Adapt tone per format: email (warm, personal), ads (punchy, concise),
  landing page (direct, confident), newsletter (casual, curious)
- Present all four outputs clearly labeled in a single response
- After each output, add a one-line note on what to verify before publishing

Step 3: Test the agent. Open a conversation in your Cowork project and invoke the agent by name. Paste in a blog post and check that you get all four outputs with the correct structure and constraints.

/skill-creator does all of this in about thirty seconds. The manual route takes five to ten minutes and gives you full control over every line of the configuration. Either path gets you the same working agent.


Run It

Time to test. Open a conversation in your Cowork project, invoke the Content Repurposer agent, and paste in the Tideway blog post about the five tax mistakes freelancers make. Here is what representative output looks like.

Email Copy

Subject line: Five tax mistakes that cost freelancers money

Preview text: Most of these are fixable in an afternoon.

Body:

You are leaving money on the table every April. Not because you are bad at your job -- because nobody taught freelancers how taxes actually work.

We see the same five mistakes every year. Mixing personal and business expenses. Missing quarterly estimated payments. Forgetting home office deductions. Ignoring mileage tracking until December. Filing as a sole proprietor when an S-corp election would save thousands.

Each one is fixable in an afternoon, and the payoff shows up the next time you file. We wrote a guide that walks through all five and tells you exactly what to do about each one.

One less thing to worry about when tax season comes around.

CTA: Read the full guide on our blog -- [5 Tax Mistakes Freelancers Make]

Verify before publishing: Confirm the S-corp election claim applies to your jurisdiction. Check that the "saves thousands" language aligns with what the blog post actually states.


Ad Headlines

  1. Five tax mistakes that cost freelancers every April
  2. Freelancers: your bookkeeping might be costing you thousands
  3. The quarterly payment mistake most freelancers make
  4. Stop losing money to tax mistakes you can fix today
  5. What your accountant wishes you knew about freelance taxes

Verify before publishing: Check character counts on your target ad platform. Confirm "thousands" is supported by the blog post content.


Landing Page Section

Hero headline: Stop overpaying on freelance taxes

Supporting paragraph: Most freelancers make the same five bookkeeping mistakes every year -- mixing accounts, missing quarterly payments, skipping deductions. Each one costs real money. Our guide breaks down what goes wrong and how to fix it, so you keep more of what you earn.

CTA button text: Get the free guide

Verify before publishing: Ensure the landing page has a way to deliver the guide (email gate, direct link, or blog redirect).


Newsletter Blurb

We published a new guide this week: the five tax mistakes we see freelancers make every single year. The one about quarterly estimated payments alone probably saves most readers a few hundred dollars in penalties. Worth ten minutes of your time.

CTA link text: Read the five mistakes

Verify before publishing: Confirm the "few hundred dollars" estimate is consistent with the source content.


That is one blog post turned into four pieces of channel-ready content. Look at the batch and ask three questions.

Does each format sound like Tideway? The email is warm and reassuring -- "one less thing to worry about" is Tideway's voice. The ad headlines are direct without being aggressive. The landing page is confident but not pushy. The newsletter is casual, like a friend sending a link. All four sound like they come from the same company.

Do the constraints hold? Check the ad headlines against the 90-character limit. Count the email body words. Make sure the landing page headline is under 10 words. These are mechanical checks -- fast to verify, easy to catch.

Does the CTA match the format? The email CTA drives to the blog. The landing page CTA drives to a guide (which could be email-gated). The newsletter CTA drives to the article. Each CTA fits the context where the reader encounters it. An email reader expects a link. A landing page visitor expects a button.

If something is off, the fix is usually specific. "The ad headlines are too soft -- make variant 3 more direct" is better feedback than "try again." Your corrections sharpen the agent for the next run.


Maintaining Voice at Scale

When you write one blog post yourself, voice consistency is automatic. The problem starts when a single piece of content becomes four derivative pieces. Each format pulls the voice in a different direction. Ads want to be punchy. Emails want to be warm. Landing pages want to be direct. If the agent optimizes for each format without anchoring to your voice, the outputs drift apart.

This is where the brand voice rules you built in Article 4 do their real work. The agent reads .claude/rules/brand-voice.md before producing anything. Those rules -- your tone adjectives, your vocabulary list, your "voice in action" examples -- act as guardrails across all four formats. The newsletter blurb should sound like the same company that wrote the email. The ad headlines should feel like the same brand that wrote the landing page section.

In practice, voice drift tends to concentrate in specific formats. Your emails might sound right while your ad headlines sound generic. That is normal -- under 90 characters leaves little room for personality, and brand voice is the first thing to get squeezed out.

The fix is format-specific voice guidance. If ads keep drifting, add a section to your brand voice rules that addresses ads directly:

## Ad Copy Voice
- Lead with the reader's problem, not our solution
- Use the same plain language as our blog -- no ad-speak
- "Your bookkeeping is costing you money" over "Optimize your financial workflows"
- Numbers are fine in ads. Vague claims are not.

You do not need format-specific guidance for every format upfront. Wait until you see drift, then tighten the rules where needed. Start loose, tighten where the output tells you to.

The repurposer agent only works with content that has already passed your Brand Voice Checker (Article 6). If the source content sounds right, the derivatives have a solid starting point. If you feed the agent a blog post that was never voice-checked, you are asking it to maintain a voice that was never established. The quality of the input sets the ceiling for the quality of the output.


What is Next

You now have four working tools. The Content Brief Generator (Article 5) produces structured briefs for new content. The Brand Voice Checker (Article 6) ensures drafts match your brand. The Campaign Strategist (Article 8) turns approved content into campaign plans. The Content Repurposer (this article) transforms approved content into email, ads, landing page sections, and newsletter blurbs.

Each one works on its own. You can run any of them independently, and they produce useful output. But right now, you are the glue. You run the brief generator, take its output, hand it to a writer (or your VP), run the voice checker on the draft, hand the approved content to the campaign strategist, then hand it to the repurposer. That is four separate steps with you manually moving content between them.

In Article 10, you wire them together into a pipeline. Brief to voice-check to campaign plan to repurposed content, end to end. You trigger the pipeline once and get finished, channel-ready output. The individual tools become a system.


This is Part 9 of 16 in the Your AI VP of Marketing series. Previous: Agent: Campaign Strategist | Next: Wiring It Together: Your First Marketing Pipeline