DEV Community: Nex Tools

How I Built a Full Content Pipeline with Claude Code (No Team, No Budget)

Nex Tools — Tue, 14 Apr 2026 16:55:28 +0000

I run a small e-commerce brand. One person. No marketing team. No content budget.

Six months ago, I was posting once a week - maybe. Now I publish daily across 5 platforms with consistent branding, scheduled posts, and automated cross-posting.

The entire system runs on Claude Code + a handful of MCP servers. Here's exactly how.

The Problem: Content Takes Forever

The math was brutal:

1 Instagram carousel: 2-3 hours (research, copy, design, caption, hashtags)
1 blog post: 3-4 hours
Cross-posting to 5 platforms: 1 hour of manual formatting

That's 6-8 hours for a single piece of content across all channels. At 5 posts per week, I'd need a full-time content person. I couldn't afford one.

The Architecture: Skills + MCP + Scripts

Claude Code's skill system is the backbone. Here's the stack:

Research Layer:    WebSearch + NotebookLM + Apify MCP
Creation Layer:    Custom skills (content-factory, copywriter, art-director)
Visual Layer:      Gemini MCP (image gen) + Puppeteer (HTML to PNG)
Publishing Layer:  Platform APIs (IG, Hashnode, DEV.to, Blogger, Medium)
Scheduling Layer:  Cron-based scheduled tasks

Each layer has specialized workers - Claude Code skills that know exactly what to do.

Step 1: Research Skills

I have research skills that scan competitors, trending topics, and viral content patterns. They output structured markdown files with hooks, angles, and data points.

The key insight: research and creation are separate skills. A researcher doesn't create. A creator doesn't research. This separation is what makes the quality consistent.

Step 2: Content Factory

A daily skill reads all research files, checks what's already been published (via a publish log), and generates:

10 quote posts (HTML with exact brand styling)
10 reel scripts (JSON with scene-by-scene timing and animations)
10 article drafts (Markdown with frontmatter for each platform)

Everything includes source attribution back to the research that inspired it.

Step 3: Visual Pipeline

Quote posts are HTML files styled with exact brand colors, fonts, and layout. A Puppeteer script converts them to PNG:

node carousel-to-png.js quotes.html output-folder

1080x1080 PNGs, ready for Instagram. No Canva. No Figma. No design skills needed.

Step 4: Platform-Specific Publishing

Each platform has its own publishing script:

Instagram: Graph API for carousels and reels
Hashnode: GraphQL API
DEV.to: REST API with canonical URLs
Blogger: Google OAuth2 + Blogger API v3
Medium: Chrome-based flow (API closed since 2023)

Step 5: Scheduling

Claude Code's scheduled tasks run the pipeline daily:

09:00 - Content Factory generates daily batch
10:00 - Publisher distributes to all platforms
11:00 - Instagram post goes live
20:00 - Inventory check + next day planning

Results After 30 Days

Before: 1 post/week, manual, inconsistent branding
After: 5+ posts/day across 5 platforms, consistent, automated
Time spent: ~30 minutes/day reviewing and approving
Cost: $0 (Claude Code subscription + free API tiers)

What I'd Do Differently

Start with one platform, not five. Should have nailed Instagram first, then expanded.
Quality gates matter. Now I have QA skills that review before publishing.
Research is the bottleneck, not creation. 14 research documents that never became content taught me this.

The Stack

Component	Tool	Cost
AI Engine	Claude Code (Opus)	Subscription
Image Gen	Gemini MCP	Free tier
HTML to PNG	Puppeteer + Node.js	Free
IG Publishing	Meta Graph API	Free
Blog Publishing	Hashnode/DEV.to/Blogger APIs	Free
Scheduling	Claude Code scheduled tasks	Included
Research	WebSearch + Apify	$5 credit

Total monthly cost: Claude Code subscription only.

Building something similar? I'm documenting the entire process. Follow for the next post on how the visual pipeline works.

10 Claude Code Skills I Wish I Had When I Launched My First Product

Nex Tools — Mon, 13 Apr 2026 22:05:26 +0000

My first product launch went like this: I spent 4 weeks building, 2 days panicking, and 1 afternoon throwing together a landing page, an email, and a prayer.

No financial model. No launch checklist. No competitor research. No outreach plan. I just shipped and hoped.

The second time, I used a system. 10 Claude Code skills that covered every step from idea validation to post-launch metrics. I launched 3 products in 30 days, and each one was more structured than anything I had shipped before.

Here is what each skill does, why the order matters, and the exact launch sequence I follow.

The Real Problem With Solo Launches

You know how to build. That is not the bottleneck. The bottleneck is everything around the building:

Validating the idea before you invest 200 hours
Knowing what your competitors charge and where they are weak
Writing a landing page that converts, not just describes
Modeling the finances so you know if the unit economics work
Planning the launch so day one is not chaos
Following up with the right people at the right time
Measuring what happened after the dust settles

Each of these tasks takes a full day when you do it manually. And most solo founders skip half of them because there is no time.

These 10 skills automate the parts that do not require creative judgment but absolutely require execution.

What Are Claude Code Skills?

If you are new here: Claude Code is Anthropic's CLI tool that puts Claude in your terminal. Skills are markdown files you drop into ~/.claude/commands/ and run with /skill-name.

No API keys. No dependencies. No configuration. Copy the file, restart, run the command. Each skill has specialized logic for a specific job.

The Launch Sequence

The order matters. Each skill produces output that feeds into the next. Here is the sequence I follow:

Day 1-3:  /mvp-builder         → scope and spec
Day 4:    /competitor-mapper    → landscape and gaps
Day 5-6:  /landing-page-gen     → conversion-focused page
Day 7:    /financial-model      → unit economics and runway
Day 8-14: /launch-checklist     → day-by-day plan
Day 14:   /outreach-sequence    → cold emails and partnerships
Day 15+:  /metrics-tracker      → KPI dashboard

The other 3 skills - User Research Synthesizer, Pitch Deck Creator, and Product Roadmap - slot in based on your situation. Raising money? Pitch Deck moves to day 5. Have survey data? User Research goes to day 1.

Let me walk through each one.

Skill 1: MVP Builder (Days 1-3)

The problem: "What is the minimum I can build that validates this idea?" sounds simple until you are 6 weeks into building features nobody asked for.

What it does:

Takes your idea description and target user
Identifies core features vs nice-to-haves using a strict MoSCoW framework
Recommends a tech stack based on your constraints (solo dev, budget, timeline)
Generates a buildable spec with acceptance criteria for each feature
Estimates development time per feature
Outputs a prioritized backlog

Example output:

Product: URL shortener with analytics
Timeline: 2 weeks solo

MUST HAVE (Week 1):
├── URL shortening (create, redirect)         - 4 hours
├── Click tracking (count, timestamp, referrer) - 6 hours
├── Basic dashboard (list links, show clicks)   - 8 hours
└── User auth (signup, login)                   - 4 hours

SHOULD HAVE (Week 2):
├── Custom aliases                              - 2 hours
├── UTM parameter passthrough                   - 3 hours
└── CSV export                                  - 2 hours

WON'T HAVE (v1):
├── Team features
├── API access
└── Custom domains

Tech stack: Next.js + Supabase + Vercel
Reason: Fastest path to production for solo dev. Free tier covers MVP traffic.

The spec is specific enough to build from. Not a vague idea doc - an actual buildable plan with time estimates.

Skill 2: Competitor Mapper (Day 4)

The problem: You know competitors exist. You do not know their exact pricing tiers, positioning language, feature gaps, or where they are losing customers.

What it does:

Maps the competitive landscape across 5-15 competitors
Extracts pricing tiers with feature comparisons
Identifies positioning gaps (what nobody is saying)
Analyzes their landing pages for messaging patterns
Finds product gaps (features users request that nobody builds)
Outputs a positioning recommendation for your product

Example output:

Competitor Grid:
┌──────────────┬────────┬──────────┬───────────────┐
│ Competitor    │ Price  │ Focus    │ Gap           │
├──────────────┼────────┼──────────┼───────────────┤
│ Bitly         │ $35/mo │ Enterprise│ Overkill for │
│               │        │          │ solo projects │
│ Short.io      │ $25/mo │ Teams    │ No free tier  │
│ TinyURL       │ Free   │ Casual   │ Zero analytics│
│ Dub.co        │ $24/mo │ Devs     │ Complex setup │
└──────────────┴────────┴──────────┴───────────────┘

Positioning gap: No one targets solo developers who want
simple analytics without enterprise pricing.

Recommended position: "Link analytics for developers.
Free tier. No team features you won't use."

This output feeds directly into the Landing Page Generator and the Financial Model. You know what to charge, who to target, and what to say.

If you are building products and doing this kind of research manually, the Solo Founder Launch Kit automates the entire research and launch process.

Skill 3: Landing Page Generator (Days 5-6)

The problem: You can build the product but the landing page takes longer than the product itself. And half the time, it describes features instead of selling outcomes.

What it does:

Generates full HTML/CSS landing pages (not wireframes, actual deployable pages)
Conversion-focused copy structure: hero, social proof, features-as-benefits, objection handling, pricing, CTA
Responsive design out of the box
Dark mode and light mode variants
Multiple CTA placements optimized for scroll depth
Outputs clean code you can host anywhere

The copy structure follows a proven pattern:

<section class="hero">
  <!-- Outcome-focused headline, not feature-focused -->
  <h1>Track every click. Zero setup. Free forever.</h1>
  <p>Link analytics for developers who ship fast
     and don't need enterprise dashboards.</p>
  <a href="#pricing" class="cta">Start Free</a>
</section>

<section class="social-proof">
  <!-- Numbers or logos -->
  <p>2,400+ links tracked this week</p>
</section>

<section class="features">
  <!-- Each feature is a user outcome -->
  <div class="feature">
    <h3>Know where your clicks come from</h3>
    <p>Referrer, country, device - no JavaScript
       snippet needed.</p>
  </div>
</section>

<section class="objections">
  <!-- Handle the "why not just use X" -->
  <h2>You don't need Bitly</h2>
  <p>If you're a solo dev tracking project links,
     you need 10% of what enterprise tools offer.
     We built that 10%.</p>
</section>

The page is ready to deploy to Vercel, Netlify, or any static host. No framework dependencies.

Skill 4: Financial Model (Day 7)

The problem: "Will this make money?" is a question most founders answer with feelings instead of math.

What it does:

Revenue projections across 3 scenarios (conservative, baseline, optimistic)
Unit economics: CAC, LTV, payback period, gross margin
Runway calculations based on your current burn rate
Breakeven analysis: how many customers at what price
Sensitivity tables: what happens if conversion drops 20%?

Example output:

BASELINE SCENARIO (Month 6):
Revenue:        $2,400/mo (120 paid users x $20/mo)
COGS:           $180/mo (hosting + API costs)
Gross margin:   92.5%
CAC:            $8.50 (content marketing + ProductHunt)
LTV:            $140 (7-month avg retention x $20)
LTV:CAC ratio:  16.5x
Breakeven:      Month 3 (at 45 paid users)

SENSITIVITY:
If conversion rate drops 30%: breakeven at month 5
If churn doubles: LTV drops to $80, still 9.4x CAC
If price drops to $15: breakeven at month 4

This is not a spreadsheet you fill in manually. The skill builds the model from your inputs and runs the scenarios automatically.

Skill 5: Launch Checklist (Days 8-14)

The problem: Launch day arrives and you realize you forgot to set up analytics, prepare your email list, write the ProductHunt tagline, or tell anyone the product exists.

What it does:

Generates a day-by-day launch plan customized to your product type (SaaS, info product, marketplace, physical product)
Pre-launch tasks: beta testers, email list warm-up, social teasers
Launch day: platform-specific submissions, email blast, social posts
Post-launch: follow-up sequences, feedback collection, iteration plan
Each task has a time estimate and dependency map

Example (SaaS launch):

DAY -7: PREPARATION
□ Set up analytics (Plausible/PostHog)           - 30 min
□ Create ProductHunt draft                        - 45 min
□ Write launch email (for existing list)          - 30 min
□ Prepare 3 social posts (teaser, launch, recap)  - 20 min
□ Line up 5 beta testers for testimonials         - 60 min

DAY -3: WARM-UP
□ Send "something's coming" email                 - 15 min
□ Post teaser on Twitter/LinkedIn                 - 10 min
□ DM 10 people who might share                    - 30 min

DAY 0: LAUNCH
□ 06:00 - Go live on ProductHunt (PST)
□ 06:30 - Send launch email
□ 07:00 - Post on Twitter, LinkedIn, HN
□ 12:00 - Respond to all comments
□ 18:00 - Thank-you post with early stats

DAY +1 to +7: FOLLOW-UP
□ Email non-openers with different subject line
□ Collect feedback from first 20 users
□ Fix top 3 reported issues
□ Write "launch retrospective" post

No more forgetting steps. No more scrambling on launch morning.

The Solo Founder Launch Kit includes this checklist generator plus 9 other skills that cover the full journey from idea to post-launch.

Skill 6: User Research Synthesizer

The problem: You have survey responses, interview notes, and support tickets. The insights are in there somewhere, buried under 50 pages of raw text.

What it does:

Ingests any research data (interview transcripts, survey CSV, support tickets, forum posts)
Identifies patterns: top pain points, feature requests, language patterns
Groups findings by user segment
Extracts exact quotes usable in marketing copy
Outputs a prioritized insight report with confidence levels

This skill is most valuable before you build. Feed it Reddit threads, competitor reviews, or survey responses from your target audience. It tells you what people actually want, in their own words.

Skill 7: Pitch Deck Creator

The problem: Investor decks follow a specific structure that is well-documented but time-consuming to execute. You know you need a problem slide, solution slide, market size, traction, and ask. Writing all of it takes days.

What it does:

Generates content for each standard slide (10-12 slides)
Problem/Solution with narrative structure
Market sizing (TAM/SAM/SOM with logic)
Business model slide with revenue mechanics
Traction slide formatted for your stage
Team slide even for solo founders (skills + advisors)
The Ask: specific amount, use of funds, milestones

This outputs slide content, not a PowerPoint file. You paste it into your template of choice. The structure and narrative are the hard part - that is what this skill handles.

Skill 8: Outreach Sequence

The problem: Cold email is a numbers game with a skill component. You need volume and you need quality. Writing 50 personalized cold emails manually is not realistic.

What it does:

Generates cold email templates for 3 audiences: press, partners, and early users
3-email sequence per audience with follow-up timing
Personalization variables for each template
Subject lines A/B tested with the same scoring as the Headline Tester
Response handling: what to say when they reply (positive, neutral, objection)

Example sequence:

EMAIL 1 (Day 0) - The opener:
Subject: "Quick question about [their product/content]"
Body: 2 sentences of genuine context + 1 sentence about
what you built + 1 clear ask (demo, feedback, share)

EMAIL 2 (Day 3) - The value add:
Subject: "Re: [original subject]"
Body: Share a specific insight or resource relevant to
them + soft reminder of your ask

EMAIL 3 (Day 7) - The breakup:
Subject: "Last one from me"
Body: Acknowledge they're busy + final CTA + leave the
door open

The templates are customized to your product and audience. Not generic "Hi {first_name}" garbage.

Skill 9: Product Roadmap

The problem: After launch, feature requests start pouring in. Without a framework for prioritization, you build whatever the loudest customer asks for.

What it does:

Organizes features into phases (Now, Next, Later)
Scores by impact and effort (2x2 matrix)
Groups by user segment (which features matter to which users)
Estimates development effort per feature
Identifies dependencies (what must ship before what)
Outputs a structured roadmap document

Example:

NOW (Sprint 1-2):
├── Custom aliases        [High impact, Low effort]  - 3 hrs
├── Click geography       [High impact, Med effort]  - 8 hrs
└── Email notifications   [Med impact, Low effort]   - 4 hrs

NEXT (Sprint 3-4):
├── API access            [High impact, High effort] - 20 hrs
├── Team sharing          [Med impact, Med effort]   - 12 hrs
└── UTM builder           [Med impact, Low effort]   - 5 hrs

LATER (Backlog):
├── Custom domains
├── A/B testing for links
└── Slack integration

This keeps you focused. When a customer asks for custom domains on week 2, you have a documented reason for why it is in the "Later" bucket.

Skill 10: Metrics Tracker

The problem: You launched. Now what? You check Stripe revenue, maybe Google Analytics, and your email open rates. But there is no unified view. No week-over-week trends. No early warning signals.

What it does:

Defines your KPI framework based on your product type and stage
Sets up tracking for: activation, retention, revenue, referral
Creates a weekly review template
Identifies leading indicators (metrics that predict outcomes before they happen)
Outputs a dashboard structure you can implement in any tool

Example framework:

NORTH STAR: Weekly Active Links Created

HEALTH METRICS:
├── Activation: % of signups who create first link (target: 60%)
├── Retention: % who create a link in week 2 (target: 40%)
├── Revenue: MRR and MRR growth rate
└── Referral: % of users from word-of-mouth

LEADING INDICATORS:
├── Links created per user per week (predicts retention)
├── Dashboard views per user (predicts activation)
└── Feature request volume (predicts churn when it spikes)

WEEKLY REVIEW:
□ Compare metrics to last week
□ Identify biggest mover (up or down)
□ One action item from the data

This is the skill you run every week after launch. It turns raw numbers into decisions.

How They Work as a System

The output chain looks like this:

MVP Builder ──→ defines what to build
     │
Competitor Mapper ──→ defines positioning
     │
Landing Page Gen ──→ uses positioning for copy
     │
Financial Model ──→ validates unit economics
     │
Launch Checklist ──→ plans execution
     │
Outreach Sequence ──→ drives launch traffic
     │
Metrics Tracker ──→ measures results
     │
Product Roadmap ──→ plans next iteration
     │
User Research ──→ validates direction
     │
Pitch Deck ──→ if raising money

Each skill produces structured output that the next one can use. The Competitor Mapper finds your positioning gap, the Landing Page Generator uses it for copy, and the Financial Model validates that the price point works.

This is not 10 random tools. It is a launch system.

Installation

Same as every Claude Code skill:

# Copy to your commands folder
cp *.md ~/.claude/commands/

# Restart Claude Code
# Run any skill
/mvp-builder

No npm install. No environment variables. No framework. The files are the skills.

Frequently Asked Questions

Is this only for SaaS products?

No. The skills adapt to your product type. I have used them for info products, physical goods, and marketplace launches. The Launch Checklist, for example, asks what you are launching and adjusts the plan accordingly. SaaS gets ProductHunt and Hacker News. Physical products get Amazon and influencer outreach. Info products get Gumroad and community seeding.

Can I use just one or two skills instead of all 10?

Absolutely. Each skill works standalone. The MVP Builder is useful even if you never run the Financial Model. The Outreach Sequence works for any cold email campaign, not just launches. Start with the skill that solves your current bottleneck.

How long does a full launch sequence take?

Following the day-by-day sequence, about 2 weeks from idea to launched product. The skills handle the research, planning, and writing. You handle the building and decision-making. Most of the time savings come from Competitor Mapper (replaces 1-2 days of manual research), Landing Page Generator (replaces 2-3 days of design and copywriting), and Launch Checklist (replaces the planning you would otherwise skip).

I am pre-revenue. Is the Financial Model useful for me?

Especially for you. The model shows you exactly how many customers you need at a given price to cover your costs. It runs scenarios so you know that if your conversion rate is 2% instead of 5%, you need 3x the traffic. Better to know that on day 7 than on day 60.

Get the Kit

The Solo Founder Launch Kit includes all 10 skills for $39. Run the launch sequence once and you will never go back to winging it.

If you also create content or run a store, the Complete Bundle includes all 22 skills across ecommerce, content creation, and product launch for $69 (saves $18 vs buying separately).

Related: Building a content engine to drive traffic to your launch? Read How I Built a Content Machine That Publishes to 4 Platforms Simultaneously for the content side of the system. Already selling products? Check out 5 Claude Code Skills That Run My Entire Ecommerce Operation.

Built by Nex Tools - we build automation tools for online businesses.

How I Built a Content Machine That Publishes to 4 Platforms Simultaneously

Nex Tools — Mon, 13 Apr 2026 21:59:45 +0000

I wrote one blog post last Tuesday. By Wednesday morning, it existed as:

A long-form SEO article on Hashnode
A Twitter thread with 11 tweets
An Instagram carousel script
A LinkedIn post with a different hook
A newsletter edition
A YouTube script with timestamps
5 more platform-native formats I did not even plan

Total time after the original article: 23 minutes.

This is not about writing faster. It is about writing once and letting a system handle the rest. Here is exactly how it works, what tools I use, and why I stopped manually reformatting content months ago.

The Problem Every Creator Knows

You publish a blog post. Now you need to promote it. So you open Twitter, stare at it, try to condense 2,000 words into 280 characters. Then you do the same for LinkedIn but with a different tone. Then Instagram. Then your newsletter.

By the time you finish reformatting one article for four platforms, it is Thursday and you have not written anything new.

The math does not work. If you publish 3 articles per week and each one needs 5 platform variations, that is 15 pieces of content. At 20 minutes each, you are spending 5 hours per week just reformatting. Not writing, not thinking, not creating - reformatting.

I automated this entire process with Claude Code skills.

What Are Claude Code Skills?

Quick primer if you are new to this: Claude Code is Anthropic's CLI tool. You work with Claude directly in your terminal. Skills are markdown files you drop into ~/.claude/commands/ that act like specialized agents.

Each skill has a specific job. You run it with /skill-name and it executes with access to your project files. No API keys. No npm install. No configuration.

Here is the system I built with 7 of them.

The Workflow: Write Once, Publish Everywhere

This is the actual sequence I run every time I write something:

Original Article
    │
    ▼
/repurpose-engine ──→ 11 platform-native outputs
    │
    ├──→ /seo-blog-writer ──→ optimized long-form version
    │
    ├──→ /social-media-formatter ──→ Twitter, LinkedIn, Instagram
    │
    ├──→ /newsletter-builder ──→ email edition
    │
    ├──→ /youtube-script-gen ──→ video script with retention hooks
    │
    └──→ /content-calendar ──→ scheduled across 30 days

Let me walk through each skill and show what it actually does.

Skill 1: Repurpose Engine - The Multiplier

This is the one that changed everything. You feed it one piece of content and it outputs 11 platform-specific versions.

What it does:

Takes any content (blog post, transcript, notes, even bullet points)
Analyzes the core message, supporting points, and emotional hooks
Outputs 11 formats: Twitter thread, LinkedIn post, Instagram caption, email snippet, YouTube script outline, podcast talking points, Reddit post, Quora answer, carousel script, short-form video hook, and blog summary

Real example: I wrote a 1,800 word article about automation for solo founders. The Repurpose Engine turned it into:

Twitter thread (11 tweets) - each tweet standalone
LinkedIn post - professional angle, different hook
Instagram caption - emotional hook + CTA
Reddit post - stripped all marketing, pure value
Email snippet - personal tone, one core insight
YouTube outline - 8 min script with retention timestamps

The key is that each output sounds native to its platform. The Twitter thread does not read like a chopped-up blog post. The LinkedIn version has a different opening, different structure, different CTA. It understands that Reddit hates self-promotion and adjusts accordingly.

Why this matters: Most repurposing tools just reformat. They take your paragraph and make it shorter. This skill rewrites for the platform's native language, attention patterns, and audience expectations.

Skill 2: SEO Blog Writer

The problem: You know your topic but your article is not ranking. The keywords are wrong, the structure is off, there is no schema markup, and your internal links point nowhere useful.

What it does:

Keyword clustering: groups related terms so you cover the topic semantically
Content gap analysis: identifies what competitors cover that you do not
Schema markup: generates JSON-LD for your article type
Internal linking suggestions based on your existing content
Meta title and description with proper character counts
H2/H3 structure optimized for featured snippets

I run this after the Repurpose Engine to optimize the long-form version. It takes my draft and turns it into something Google actually wants to show people.

Before and after:

Before: "How to Use AI Tools for Content"
After:  "AI Content Pipeline: How to Create 11 Posts From One Article (2026 Workflow)"

The second version targets a long-tail keyword with actual search volume, includes a number for CTR, and has a year tag for freshness signals.

If you are publishing content regularly and want a system that handles the optimization side, the Content Creator Toolkit includes this skill plus six others that chain together.

Skill 3: YouTube Script Generator

The problem: Writing for video is different from writing for text. The first 30 seconds determine whether anyone watches. Most blog-to-video conversions lose viewers in the first 10 seconds because the opening was written for readers, not watchers.

What it does:

Generates retention-architected scripts (hook, loop, payoff structure)
Timestamps with topic markers
Thumbnail concepts (text overlay + emotion + visual hook)
B-roll suggestions at each section break
CTAs placed at retention peaks, not at the end when nobody is watching
Estimated runtime based on speaking pace

The output looks like this:

[0:00-0:30] HOOK
"I wrote one article last week. By Friday, it was live on 4 platforms,
 got 2,400 views, and I spent 23 minutes total after the first draft."
[Cut to: screen recording of terminal running /repurpose-engine]

[0:30-2:00] CONTEXT
Why most creators burn out: the reformatting trap.
[B-roll: split screen - 4 platform tabs open, typing the same thing]

[2:00-5:30] THE SYSTEM
Walk through the 7 skills, show terminal output for each.
[Screen recording: actual Claude Code session]
...

Every section has a purpose. The hook creates curiosity. The context builds relevance. The walkthrough delivers value. CTAs appear at 30%, 60%, and at the end.

Skill 4: Headline A/B Tester

The problem: You spend 3 hours writing an article and 5 seconds on the headline. The headline determines whether anyone reads the 3 hours of work.

What it does:

Generates 10+ headline variants from your content
Scores each across 7 dimensions: clarity, curiosity, specificity, emotion, SEO value, social shareability, platform fit
Adjusts scores per platform (what works on Twitter fails on LinkedIn)
Ranks them with reasoning
Suggests a "safe pick" and a "swing pick"

Example scoring:

"How I Built a Content Machine That Publishes to 4 Platforms Simultaneously"
├── Clarity: 9/10 - immediately clear what you learn
├── Curiosity: 7/10 - "how" creates mild curiosity
├── Specificity: 8/10 - "4 platforms" is concrete
├── Emotion: 5/10 - functional, not emotional
├── SEO: 6/10 - no primary keyword
├── Social: 7/10 - shareable result
└── Platform (Hashnode): 8/10 - dev audience likes "how I built"

Overall: 7.1/10 - Strong for technical audience

I run this for every piece of content. It takes 30 seconds and consistently produces better headlines than what I come up with on my own.

Skill 5: Content Calendar

The problem: Publishing randomly is not a strategy. But building a content calendar manually means staring at a spreadsheet for an hour every month.

What it does:

Builds a 30-day publishing plan from your topic list
Weights content by pillar (70% core topic, 20% adjacent, 10% experimental)
Assigns KPIs per post (traffic, engagement, conversion)
Accounts for platform-specific timing (Tuesday for blogs, Thursday for LinkedIn)
Includes repurposing schedule (original on day 1, variations across days 2-5)

Output looks like:

Week 1:
  Mon: [Blog] "AI Content Pipeline" → Hashnode
  Tue: [Thread] Repurposed from blog → Twitter
  Wed: [Carousel] 7 skills breakdown → Instagram
  Thu: [Post] "Why I stopped reformatting" → LinkedIn
  Fri: [Newsletter] Weekly edition → Email list
  KPIs: 2,000 blog views, 15 newsletter signups

One article feeds the entire week. That is the system working.

Want to see all 7 skills working together? The Content Creator Toolkit is the complete set.

Skill 6: Newsletter Builder

The problem: Your newsletter is supposed to go out every week. Half the time, you skip it because "you do not have anything to say" - even though you published 3 pieces of content that week.

What it does:

Pulls from your existing content (blog posts, social posts, notes)
Structures a complete newsletter edition with sections
Writes a subject line with open rate optimization
Includes a personal intro, curated links, and a CTA
Matches your newsletter's voice and format

The structure it generates:

Subject: "The 23-minute content trick (and why I stopped using Notion)"

Hey [first name],

This week I accidentally discovered something about content repurposing
that saved me 5 hours. Let me show you.

[Section 1: Main insight from this week's blog post]
[Section 2: Quick tip from social posts]
[Section 3: What I'm reading/using]
[CTA: Link to full article or product]

No more staring at a blank email draft on Friday afternoon.

Skill 7: Social Media Formatter

The problem: Every platform has its own rules. Character limits, hashtag strategies, image ratios, link placement, emoji conventions. Keeping track of all of it while trying to write good content is exhausting.

What it does:

Takes any source material (article, notes, bullet points)
Outputs platform-native posts for Twitter, LinkedIn, Instagram, Facebook, Reddit, and Threads
Handles character limits, hashtag research, and CTA placement
Adjusts tone per platform (professional for LinkedIn, casual for Twitter, visual for Instagram)
Includes posting notes (best time, hashtag count, image suggestions)

The Formatter is the last step in the chain. Everything the Repurpose Engine generates gets polished here for final publishing.

The Full Chain in Action

Here is what a typical content day looks like now:

9:00 AM - Write one article (60-90 minutes)
10:30 AM - Run /repurpose-engine (2 minutes)
10:35 AM - Run /seo-blog-writer on the long-form version (3 minutes)
10:40 AM - Run /headline-ab-tester on all versions (1 minute)
10:42 AM - Run /social-media-formatter on social versions (2 minutes)
10:45 AM - Run /newsletter-builder for the weekly edition (3 minutes)
10:50 AM - Run /content-calendar to slot everything (2 minutes)

Total: 90 minutes writing + 13 minutes processing = one article becomes content for the entire week.

Compare that to writing each piece separately. 5 platforms, 20 minutes each, 3 articles per week. That is 5 hours saved every single week.

Installation

Every skill is a single .md file:

# Copy files to your Claude Code commands folder
cp *.md ~/.claude/commands/

# Restart Claude Code
# Run any skill
/repurpose-engine

No dependencies. No environment variables. No build steps. The markdown file IS the skill.

Frequently Asked Questions

Can I use these skills for any niche or are they marketing-specific?

They work for any topic. The skills include niche calibration, meaning they ask about your audience and adjust output accordingly. I have seen people use them for developer tutorials, fitness content, finance newsletters, and SaaS product updates. The platform formatting rules are universal.

Do the skills work together or do I need to manually copy output between them?

They work both ways. Each skill can run standalone for a specific task, or you can chain them. The output from the Repurpose Engine becomes the input for the Social Media Formatter. The SEO Blog Writer output feeds into the Headline A/B Tester. You just paste or pipe the output from one into the next.

How is this different from using ChatGPT or Claude directly?

Three differences. First, skills run in your terminal with access to your project files, so they can read your existing content and output to files. Second, each skill has specialized logic built in - the YouTube Script Generator understands retention architecture, the SEO Blog Writer understands schema markup. Third, they are repeatable. You run the same skill every week and get consistent output format, not a random chat response.

What if I only publish on one or two platforms?

Start with the Repurpose Engine and the platform-specific skills you need. The Content Calendar is useful even for single-platform creators because it handles scheduling and pillar weighting. You do not need to use all 7 on day one.

Get the Toolkit

The Content Creator Toolkit includes all 7 skills for $29. Drop them in, run them, and stop spending hours reformatting content that could be automated.

If you also run a store or are launching a product, the Complete Bundle includes all 22 skills across ecommerce, content, and product launch for $69 (saves $18 vs buying separately).

Related: If you sell products online, check out 5 Claude Code Skills That Run My Entire Ecommerce Operation - same approach, different problem.

Built by Nex Tools - we build automation tools for online businesses.

5 Claude Code Skills That Run My Entire Ecommerce Operation

Nex Tools — Mon, 13 Apr 2026 21:59:30 +0000

If you run a Shopify store (or any ecommerce), you know the grind: rewriting product pages, building email sequences, analyzing competitors, digging through reviews, recalculating margins every time a cost changes.

I automated all of it with Claude Code skills. Five markdown files in a folder. No API keys, no config, no dependencies. Just drop them in and run.

Here is what each one does and why it matters.

What Are Claude Code Skills?

Claude Code is Anthropic's CLI tool that lets you work with Claude directly in your terminal. Skills are markdown files you drop into .claude/commands/ that act like specialized agents - each one trained for a specific job.

You run them with /skill-name and they execute. No setup beyond copying the file.

Skill 1: Product Page Optimizer

The problem: Most product pages are written once and never touched. The title is whatever you typed at 2 AM. The description talks about features nobody cares about. The SEO meta tags are either missing or auto-generated garbage.

What it does:

Rewrites product titles for both humans and search engines
Generates benefit-driven descriptions (not feature lists)
Creates SEO meta titles and descriptions with proper character counts
Suggests image alt text based on the product
Outputs everything in a format you can paste straight into Shopify

Real result: I ran this on our top 5 products. Average time on page went up 22% in the first week.

Skill 2: Email Sequence Builder

The problem: Every ecommerce store needs at least three email flows: welcome, abandoned cart, and win-back. Most stores have zero or one.

What it does:

Generates complete email sequences with subject lines, body copy, and send timing
Welcome series (5 emails over 14 days)
Abandoned cart recovery (3 emails over 72 hours)
Win-back campaign (4 emails over 30 days)
Adapts tone and offers to your specific product and audience

Real result: Our abandoned cart flow recovers about 12% of carts. Before this skill, we had no flow at all.

Skill 3: Competitor Analysis

The problem: You know your competitors exist. You do not know their exact pricing, positioning, or where they are weak.

What it does:

Maps competitor pricing across product lines
Analyzes their messaging and unique selling points
Identifies gaps in their product catalog
Compares shipping, return policies, and trust signals
Outputs a structured comparison table

Why it matters: I found that none of our competitors offered a bundle discount. We added one and it became 30% of revenue in the first month.

Skill 4: Review Analyzer

The problem: Customer reviews contain gold - objections, use cases, language patterns. But reading 200 reviews manually is not happening.

What it does:

Ingests reviews from any source (CSV, text, or pasted)
Identifies top pain points, praise patterns, and objections
Extracts exact customer language for ad copy
Flags product issues before they become trends
Generates "voice of customer" snippets for landing pages

Real result: We pulled the phrase "finally something that actually works" from reviews and used it as a headline. CTR went up 40%.

Skill 5: Pricing Strategy

The problem: "Should I raise prices?" is the question every founder asks and nobody answers with data.

What it does:

Models margin scenarios across multiple price points
Calculates breakeven at each price with your actual COGS
Simulates volume changes (if price goes up 20%, how much volume can you lose?)
Compares against competitor pricing from Skill 3
Outputs a recommendation with confidence level

Why it matters: We raised prices 15% based on this analysis and lost zero customers. That is pure margin.

How They Work Together

The output of one skill feeds into the next:

Competitor Analysis → spots pricing gap
Pricing Strategy → models the opportunity
Product Page Optimizer → rewrites pages for new positioning
Email Sequence Builder → creates launch flow for new pricing
Review Analyzer → tracks customer response

This is not five separate tools. It is a system.

Getting Started

Each skill is a single .md file. Installation:

Copy the files to ~/.claude/commands/
Restart Claude Code
Run /skill-name in any project

No npm install. No environment variables. No configuration files.

Frequently Asked Questions

Do I need a paid Claude subscription?

Yes - Claude Code requires an Anthropic API key or a Claude Pro/Team subscription.

Do these work with WooCommerce or other platforms?

Yes. The skills generate text output. They work with any ecommerce platform where you can paste or import text - Shopify, WooCommerce, BigCommerce, Wix, Squarespace.

Can I customize the skills?

Every skill is a markdown file. You can edit it, add industry-specific context, or change the output format. The file IS the configuration.

How is this different from ChatGPT prompts?

Claude Code skills run in your terminal with access to your project files. They can read your existing product data, analyze your codebase, and output directly to files. Regular chat prompts cannot do that.

Get the Pack

All 5 skills are available as the Ecommerce Growth Pack ($19). Drop them in, run them, and see results the same day.

If you also create content or are launching a product, the Complete Bundle includes 22 skills across ecommerce, content creation, and product launch ($69 - saves $18 vs buying separately).

Built by Nex Tools - we build automation tools for online businesses.