SEO Machine: The Claude Code Workspace Killing Manual Content Workflows

SEO Machine: The Claude Code Workspace Killing Manual Content Workflows

Your content team is bleeding money. Every 3,000-word article costs 15+ hours of research, writing, editing, and optimization. Freelancers charge $500–$2,000 per post. Your competitors publish twice as fast with half the headcount. Meanwhile, you're stuck in Google Docs hell, copying SERP data into spreadsheets, manually checking keyword density, and praying your writer remembered internal links.

What if a single command could slash that to under 2 hours?

Meet SEO Machine—a specialized Claude Code workspace that transforms AI from a chatbot into a fully autonomous content engine. Built by Craig Hewitt (founder of podcast hosting platform Castos) and now open-sourced on GitHub, this system doesn't just write articles. It researches competitors, analyzes search intent, optimizes for featured snippets, scores your content quality 0-100, and even publishes directly to WordPress^{↗ Bright Coding Blog} with Yoast SEO metadata.

This isn't another AI writing tool. It's a complete content operations system running inside Claude Code. And it's about to make your old workflow look like a typewriter.

What Is SEO Machine?

SEO Machine is a Claude Code workspace—a structured environment of custom commands, specialized AI agents, marketing skills, and Python^{↗ Bright Coding Blog} analysis modules designed specifically for long-form, SEO-optimized content creation. Originally developed to power Castos's content marketing^{↗ Bright Coding Blog}, Hewitt open-sourced it so any business can replicate their production system.

The repository lives at github.com/TheCraigHewitt/seomachine and installs in minutes. But beneath that simplicity lies a sophisticated architecture:

• 19 custom slash commands orchestrating entire workflows from research to publication
• 10 specialized AI agents that automatically analyze and improve every piece of content
• 26 marketing skills covering copywriting^{↗ Bright Coding Blog}, CRO, A/B testing, email sequences, and pricing strategy
• 22 Python modules for real-time data integration and advanced SEO analysis
• 8 context files that lock in your brand voice, style guide, and competitive positioning

What makes SEO Machine genuinely different from Jasper, Copy.ai, or ChatGPT? Context persistence and workflow automation. Most AI tools forget your brand between sessions. SEO Machine embeds your entire content strategy—voice, examples, keywords, competitors, internal link map—into every command execution. The agents don't just generate text; they execute your strategy with machine precision.

The timing is critical. Google's March 2024 core update crushed thin AI content. SEO Machine's built-in quality scoring, readability analysis, and competitive benchmarking ensure you publish what actually ranks: comprehensive, human-sounding, data-driven content that serves search intent.

Key Features That Separate Pros from Amateurs

Command-Driven Workflow Engine

SEO Machine replaces scattered tools with unified commands. Type /research content marketing strategies for B2B SaaS and watch the system:

• Scrape and analyze top 10 SERP competitors
• Identify content gaps your competitors missed
• Generate a complete research brief with recommended outline
• Save everything to /research/ with automatic dating

No more jumping between Ahrefs, Google Docs, and spreadsheets. The command structure enforces discipline: research before writing, optimize before publishing.

Autonomous Agent Swarm

After /write completes a 2,500-word draft, four agents automatically fire:

The Content Analyzer (newly added) runs five Python modules for surgical precision: search intent classification, keyword clustering with TF-IDF and K-means, content length benchmarking against SERP competitors, multi-metric readability scoring, and comprehensive SEO quality rating with publishing readiness assessment.

Real-Time Data Integration

Most content tools operate in a vacuum. SEO Machine connects to Google Analytics 4, Google Search Console, and DataForSEO for live performance insights. The /priorities command uses this data to build a content prioritization matrix—identifying position 11-20 quick wins, declining content needing refresh, and low-CTR opportunities with actual traffic impact estimates.

Landing Page & CRO Specialization

Beyond blog content, dedicated commands handle conversion-optimized landing pages:

• /landing-research analyzes competitor positioning
• /landing-write generates pages with above-fold optimization
• /landing-audit scores CRO elements 0-100 with category breakdowns
• Six Python modules analyze headline effectiveness, CTA distribution, trust signals, and friction points

Use Cases: Where SEO Machine Dominates

1. Scaling Content for Bootstrapped SaaS

You're a 5-person team competing against VC-backed giants pumping 50 articles monthly. SEO Machine lets one marketer produce research-backed, optimized 3,000-word articles in 90 minutes—complete with internal linking strategy and meta elements. The Castos example in examples/castos/ proves this: a podcast hosting company building topical authority without a content agency.

2. Refreshing Stagnant Content Libraries

That 200-post blog from 2019? It's a goldmine hiding in plain sight. /analyze-existing fetches any URL, scores content health 0-100, identifies outdated statistics, assesses competitive positioning, and generates a rewrite priority queue. One command replaces your manual content audit spreadsheet.

3. Agency Client Onboarding at Scale

SEO agencies waste hours understanding new client voices. With SEO Machine, you populate 8 context files once—brand voice, writing examples, style guide, competitor analysis—and every subsequent article automatically adheres. Client-specific agents ensure consistency that freelancers can't match.

4. Data-Driven Editorial Calendar Building

Stop guessing what to write next. The research commands—/research-gaps, /research-trending, /research-performance—combine competitive intelligence with your own analytics to surface high-opportunity topics. The opportunity_scorer.py module calculates 8-factor scores for prioritization.

Step-by-Step Installation & Setup Guide

Prerequisites

• Claude Code installed (Anthropic's CLI tool)
• Active Anthropic API account
• Python 3.8+ for analysis modules

Installation

Clone the repository and enter the directory:

git clone https://github.com/TheCraigHewitt/seomachine.git
cd seomachine

Install Python dependencies for data integrations and analysis:

pip install -r data_sources/requirements.txt

This single command installs the entire analytics stack:

• Google Analytics 4 and Search Console API clients
• DataForSEO competitive intelligence client
• NLP libraries (nltk, textstat) for readability analysis
• scikit-learn for machine learning clustering
• beautifulsoup4 for web scraping competitor content

Launch Claude Code in the workspace:

claude-code .

Critical: Configure Your Context Files

This step determines success or failure. All files in context/ are templates—fill them with your company's specifics:

Quick Start Hack: Copy the complete Castos example from examples/castos/ as your structural template, then replace content with your specifics.

Optional: Analytics & WordPress Integration

For live data and direct publishing:

1. Copy config/competitors.example.json to config/competitors.json and customize
2. Add API credentials to .env (see data_sources/README.md)
3. Install the WordPress MU-plugin from wordpress/seo-machine-yoast-rest.php
4. Configure WordPress credentials in .env:

WP_URL=https://yoursite.com
WP_USERNAME=your_username
WP_APP_PASSWORD=your_application_password

REAL Code Examples from the Repository

Example 1: Complete New Content Workflow

This is the exact workflow from SEO Machine's documentation, executed entirely through custom commands:

# Step 1: Research before writing—non-negotiable
/research content marketing strategies for B2B SaaS
# Output: /research/brief-content-marketing-strategies-2025-10-29.md
# Contains: primary/secondary keywords, top 10 competitor analysis,
# content gaps, recommended outline, internal linking strategy

# Step 2: Generate full article with automatic agent execution
/write content marketing strategies for B2B SaaS
# Output: /drafts/content-marketing-strategies-2025-10-29.md
# Auto-triggers: SEO Optimizer, Meta Creator, Internal Linker, Keyword Mapper
# Each agent writes its analysis report to /drafts/ for your review

# Step 3: Final polish with comprehensive audit
/optimize drafts/content-marketing-strategies-2025-10-29.md
# Output: optimization report with SEO score 0-100,
# priority fixes, quick wins, publishing readiness assessment

# Step 4: Publish directly to WordPress (optional)
/publish-draft drafts/content-marketing-strategies-2025-10-29.md
# Pushes article with Yoast SEO metadata via REST API

Why this matters: Traditional workflows require 5+ tools (Ahrefs, Google Docs, SurferSEO, WordPress admin). SEO Machine compresses this into four commands within a single terminal window, with agents handling analysis that humans previously did manually.

Example 2: Content Refresh Pipeline

Existing content decay is real—this workflow combats it systematically:

# Analyze any URL or local file for improvement opportunities
/analyze-existing https://yoursite.com/blog/marketing-guide
# Alternative: /analyze-existing published/marketing-guide-2024-01-15.md

# Output saved to /research/analysis-marketing-guide-2025-10-29.md
# Includes:
# - Content health score (0-100)
# - Quick wins (immediate improvements)
# - Strategic improvements (long-term positioning)
# - Rewrite priority: Light Update / Moderate Refresh / Full Rewrite
# - Research brief pre-generated for rewrite command

# Execute rewrite based on analysis findings
/rewrite marketing guide
# Output: /rewrites/marketing-guide-rewrite-2025-10-29.md
# Contains: rewritten content, change summary, before/after comparison,
# updated SEO elements, preserved effective sections

# Final optimization pass
/optimize rewrites/marketing-guide-rewrite-2025-10-29.md

The intelligence: Unlike blind rewriting, this preserves what already works. The analysis identifies specifically which sections rank well and which decayed, preventing the common mistake of destroying effective content during "improvements."

Example 3: Python Research Scripts for Power Users

For analysts who want direct data access, SEO Machine exposes its research engine via command-line scripts:

# Run from repository root—no Claude Code needed for data gathering

# Identify quick-win opportunities (position 11-20 keywords)
python3 research_quick_wins.py

# Map competitor content gaps for strategic planning
python3 research_competitor_gaps.py

# Build performance-based content priority matrix
python3 research_performance_matrix.py

# Comprehensive prioritization with 8-factor opportunity scoring
python3 research_priorities_comprehensive.py

# Deep SERP analysis for specific keyword targeting
python3 research_serp_analysis.py

# Discover topic cluster opportunities
python3 research_topic_clusters.py

# Surface trending topics before competitors
python3 research_trending.py

# SEO baseline and competitor tracking (requires config/competitors.json)
python3 seo_baseline_analysis.py
python3 seo_competitor_analysis.py

# Verify DataForSEO API connectivity
python3 test_dataforseo.py

Technical depth: These scripts leverage the same Python modules that power the Claude Code agents—search_intent_analyzer.py, keyword_analyzer.py, content_length_comparator.py—but expose them for batch processing, cron jobs, or integration with external BI tools.

Example 4: Content Analyzer Agent Deep Dive

The newest agent demonstrates SEO Machine's analytical sophistication:

Content Analyzer Agent Execution (triggered automatically or manually)

Input: Any article file or URL

Processing Pipeline:
1. search_intent_analyzer.py
   - Classifies: Informational / Navigational / Transactional / Commercial
   - Analyzes SERP features for target keyword
   - Confidence score + content alignment recommendations

2. keyword_analyzer.py
   - Exact keyword density calculation
   - Keyword stuffing risk detection with warnings
   - Topic clustering via TF-IDF + K-means
   - Distribution heatmap by section (intro, H2s, body, conclusion)
   - LSI (Latent Semantic Indexing) keyword identification

3. content_length_comparator.py
   - Fetches top 10-20 SERP competitors
   - Calculates median, 75th percentile word counts
   - Shows your competitive positioning gap
   - Data-driven expansion recommendations

4. readability_scorer.py
   - Flesch Reading Ease score
   - Flesch-Kincaid Grade Level
   - Sentence complexity analysis
   - Passive voice ratio
   - Transition word usage
   - Overall readability 0-100

5. seo_quality_rater.py
   - Comprehensive SEO score 0-100
   - Category breakdowns: content, keywords, meta, structure, links, readability
   - Critical issues / warnings / suggestions
   - Publishing readiness determination

Output: Executive summary with priority action plan, competitive positioning,
exact metrics, and specific revision recommendations per analysis area

Why this beats manual auditing: A human SEO specialist takes 45-60 minutes for this depth of analysis. The agent completes it in under 60 seconds with consistent methodology and quantified benchmarks.

Advanced Usage & Best Practices

Maximizing Agent Effectiveness

Don't skip agent reports. After /write, four agents generate recommendations. Read them sequentially: SEO Optimizer first for structural fixes, then Keyword Mapper for density adjustments, Internal Linker for link placement, Meta Creator last for title options. Each builds on the previous.

The /scrub Command: AI Detection Evasion

Google's algorithms increasingly flag AI-written content. SEO Machine's /scrub command removes telltale patterns:

• Excessive em-dashes (AI favorite)
• Filler phrases ("it's important to note," "in today's world")
• Repetitive sentence structures
• Robotic transitions

Best practice: Run /scrub after /write, then manually review for voice consistency.

Context File Maintenance Schedule

Landing Page Optimization Workflow

For conversion-focused pages, use the specialized command chain:

/landing-research [topic]      # Competitor and positioning research
/landing-write [topic]         # Generate conversion-optimized page
/landing-audit [file]          # CRO scoring with specific fixes
/landing-competitor [URL]      # Deep competitor page teardown
/landing-publish [file]        # Push to WordPress

The CRO Analyst and Landing Page Optimizer agents analyze above-fold effectiveness, CTA quality and distribution, trust signal presence, and friction points—elements blog-focused tools ignore.

Comparison with Alternatives

SEO Machine wins where integration matters. Individual tools may exceed it in single dimensions—SurferSEO's content editor is slicker, Jasper's templates are more polished—but none unify research, creation, analysis, optimization, and publication in a single extensible system.

FAQ

Q: Do I need coding skills to use SEO Machine? A: Basic command-line familiarity helps for installation, but daily usage is pure natural language commands (/research, /write). Python scripts are optional power-user features.

Q: How does this compare to using ChatGPT or Claude directly? A: Raw LLMs forget context between sessions and require repetitive prompting. SEO Machine persists your entire content strategy across sessions and enforces workflow discipline through command structure.

Q: Will Google penalize AI-generated content from SEO Machine? A: The system is designed for quality, not volume. Built-in readability scoring, competitive benchmarking, and the /scrub command specifically target human-sounding output. Google's issue is thin content—SEO Machine's 2,000-3,000+ word minimum with unique research addresses this.

Q: Can I use this with my existing content team? A: Absolutely. SEO Machine augments human writers rather than replacing them. Writers focus on strategy and refinement while agents handle mechanical optimization.

Q: What are the ongoing API costs? A: Claude Code usage via Anthropic API, plus optional DataForSEO credits. Most users spend $20-50/month on API calls versus $500+ for premium SEO tool subscriptions.

Q: How do I customize agents for my industry? A: Agent definitions live in .claude/agents/ as markdown^{↗ Smart Converter} files. Edit prompts directly—no coding required. The repository structure encourages forking and customization.

Q: Is WordPress required? A: No. /publish-draft is optional; files remain in /drafts/ or /published/ for any CMS. The WordPress integration is a convenience, not a dependency.

Conclusion: The Future of Content Operations Is Here

SEO Machine represents a fundamental shift in how technical teams approach content marketing. It's not a writing tool—it's an operating system for organic growth, embedding strategy into executable commands and transforming AI from a creative assistant into a production engine.

The open-source model is critical. Unlike black-box SaaS tools, you own your workflow, can audit every agent's logic, and extend functionality for your specific needs. Craig Hewitt built this to scale Castos without a content agency; now any team can replicate that advantage.

My assessment after deep analysis: SEO Machine isn't perfect—the UI is terminal-based, context setup requires upfront investment, and you'll need to iterate on voice calibration. But for teams serious about systematic content production, nothing else matches this integration depth at any price point.

Your move: Clone SEO Machine from GitHub, configure your context files using the Castos example as your blueprint, and run /research [your highest-priority topic]. Within 24 hours, you'll have a publish-ready article that outperforms 90% of what's currently ranking.

The content teams that adopt systems like this in 2024 will dominate organic search in 2025. The rest will wonder why their manual workflows suddenly stopped working.

Start building your content machine today. 🚀