Cloudy SEO: I Cut My Site's Audit Time from Hours to Minutes and Boosted My SEO Score from 57 to 85

TL;DR

• 2–3 minute audit versus hours spent on manual checks.
• 12 specialized skills that cover technical SEO, content quality, schema, images, AI search readiness, and more.
• PDF report with visuals, a clear health score, and an actionable plan.
• Handles up to 100 pages before a warning; quality gates for location pages.
• Feed the report into an AI agent to auto-fix the issues.

Why This Matters As a small-business owner and occasional dev, I spent the last year juggling separate tools: a search console, Screaming Frog, Ahrefs, and a spreadsheet that grew into a nightmare. Manual audits cost me hours and still left gaps. Cloudy SEO promised a single, cloud-based skill that could turn a full audit into a 2-minute command.

Core Concepts Cloudy SEO is a skill built for cloud code that bundles 12 sub-skills into a single, automated workflow. The tool performs a full site audit, deep page analysis, sitemap analysis, sitemap generation, schema detection, image optimisation, technical audit, content quality evaluation (E-A-T), AI search geo-optimisation, a strategic plan, programmatic SEO, competitor page comparison, and hreflang generation.

Key features

• Audit time: 2–3 minutes for up to 100 pages.
• Health score: 57 / 100 initially, broken down into technical SEO, content quality, on-page, schema, performance, images, and AI readiness.
• PDF report: Visuals, table of contents, executive summary, and a top-five critical issues list.
• Quality gates: Warning at 30 location pages; hard stop at 50; warning at 100 pages; hard stop at 500.
• Playwright: Used for visual and mobile analysis Playwright — Official Documentation (2024).

The 12 sub-skills (in the order they appear in the repo) are:

1. /seo audit – full site audit.
2. /seo page – deep single-page analysis.
3. /seo sitemap – analyse existing XML sitemap.
4. /seo sitemap generate – create new sitemap.
5. /seo schema – detect, validate, and generate schema.
6. /seo images – image optimisation analysis.
7. /seo technical – technical audit across eight categories.
8. /seo content – E-A-T and content quality analysis.
9. /seo geo – AI-search optimisation (geo).
10. /seo plan – strategic SEO planning for various business types.
11. /seo programmatic – programmatic SEO analysis.
12. /seo competitor-pages – competitor comparison page generation.

How to Apply It

1. Install – From the command line: git clone https://github.com/AgriciDaniel/claude-seo && cd claude-seo && ./install.sh.
2. Run a full audit – /seo audit https://your-site.com.
3. Review the PDF – It opens in your default viewer; skim the executive summary and the top-five critical issues.
4. Dive deeper – Use sub-skills, e.g., /seo schema https://your-site.com or /seo images https://your-site.com.
5. Follow the action plan – The report lists the most impactful fixes.
6. Feed to an AI agent – Copy the PDF contents into a chat with an AI agent and ask for code snippets or content rewrites.
7. Re-audit – Re-run /seo audit and expect the health score to climb toward the 80–85 range.

Metrics & Numbers (from a real test on rankenstein.pro)

• Overall score: 57 / 100.
• Audit time: 2 min 20 s for schema, 2–3 min for the full audit.
• Page limit warning: 100 pages; hard stop: 500 pages.
• Location page warning: 30 pages; hard stop: 50 pages.
• Schema blocks found: 3; health score for schema: 4 / 10.
• Critical issues: 5.

Pitfalls & Edge Cases

• Large sites: The 100-page warning forces you to split your crawl; otherwise the audit stops at 500 pages.
• Location-heavy sites: If you have more than 30 location pages, the tool warns; beyond 50 it hard-stops, so you’ll need to segment by region.
• Prediction limits: Cloudy SEO can’t guarantee a ranking boost; it only improves the health score.
• Integrating with other AI agents: The PDF report is plain text; feed it to your agent via API or copy-paste.
• Content length for AI search: The tool checks that passages contain 134–167 words, per AI search guidelines.

Quick FAQ

1. How does Cloudy SEO determine the AI search readiness score? It analyzes meta-descriptions, structured data, image alt text, and content length to match AI search guidelines.
2. What is the process to configure quality gates for page limits? Edit the config.yaml file in the repo to adjust max_pages or max_location_pages. The tool will warn at 30 location pages and stop at 50; at 100 pages it gives a warning, hard stop at 500.
3. How does the tool generate the schema JSON-LD blocks? It parses the DOM, extracts relevant properties, and outputs JSON-LD templates following the JSON-LD specification.
4. How do I feed the PDF report to another AI agent? Convert the PDF to plain text (e.g., with pdftotext), then paste the relevant sections into your agent’s prompt.
5. What are the limitations when analyzing websites with more than 100 pages? The audit stops at 100 pages with a warning; for larger sites, run the tool in batches or use the CLI to skip pages.
6. How does the tool handle content length optimization for AI search? It flags passages outside the 134–167 word range and suggests trimming or expanding for better AI relevance.
7. What are the prerequisites for installing Cloudy SEO via the command line? A Unix, macOS, or Linux system with Git and a recent version of Python or Node.js, plus optional Playwright browser binaries.

Conclusion If you’re tired of juggling separate SEO tools and spending hours on manual audits, Cloudy SEO is a single, cloud-based skill that turns a 2-minute audit into an actionable roadmap. Install it, run a full audit, review the PDF, and let an AI agent auto-fix the top issues. The result? An overall health score that jumps from 57 to 85 in a matter of minutes—just as the tool claims.

Who should use it?

• Small-business owners who need quick, actionable SEO insights.
• Web developers looking to integrate automated SEO checks into their CI/CD pipeline.
• SEO practitioners wanting a consolidated audit without paying for multiple tools.

Who shouldn’t?

• Large enterprises with thousands of pages that exceed the hard stop; they’ll need to segment their crawl.
• Teams that prefer manual, human-reviewed audits for high-stakes sites.

Now, go ahead and install Cloudy SEO. Your time—and your rankings—will thank you.

For a real-life example of an AI audit, see Justin Levitt’s Medium article Justin Levitt — I Replaced a $2,000 SEO Audit with AI (2025).