Claude SEO Team: The Complete SE Ranking Skills Playbook

Source: github.com/seranking/seo-skills (MIT licence, 100 stars, 23 forks, official repo built by SE Ranking themselves). Production Claude Agent Skills powered by the SE Ranking remote MCP server.

26 skills covering content briefs, AI Search share of voice, technical audits, backlink gaps, keyword clusters, schema, sitemap health, GEO, and more. Each skill is a single SKILL.md file that turns raw SE Ranking API data into a finished SEO deliverable, not just a data dump.

The skills are built for the SE Ranking MCP specifically but document every API call explicitly, so they can be adapted to other SEO data providers if needed. 25 of the 26 skills come with a real, dated, publicly-run example output (against wix.com, linear.app, stripe.com, vercel.com, hostinger.com, and others) in the repo's examples/ folder, so you can see exactly what a finished run looks like before installing anything.

How a skill run actually works

Each skill is a single SKILL.md file with YAML frontmatter. Claude loads the frontmatter at startup to know when to trigger the skill, loads the body only once triggered, and reads any bundled reference files only when actually needed. This is Anthropic's progressive disclosure pattern, the same principle behind every well-built skill collection in this workspace.

A typical run produces three things: a set of numbered raw-data markdown files (01-domain-overview.md, 02-competitors.md, and so on) for auditability, optionally a CSV for spreadsheet work, and a final synthesised deliverable (BRIEF.md, REPORT.md, or PLAN.md) written in a format a human reviewer or a client can consume directly, not raw API output.

Every skill writes its output to a folder named {skill-slug}-{target-slug}-{YYYYMMDD}/, so re-runs are non-destructive and comparing output over time is a simple folder diff.

Four real install methods from the README, in order of ease.

Method 1: Claude Desktop (recommended, no terminal)

1. Open Customize in the sidebar.
2. Click Personal plugin, then + Create plugin.
3. Click Add marketplace and enter seranking/seo-skills.
4. Install the plugin once the marketplace loads.

Skills are available in your next Cowork session. Trigger by name (seo-content-brief, seo-page) or just describe what you want and Claude picks the right skill.

Method 2: Claude Code (slash commands)

/plugin marketplace add seranking/seo-skills
/plugin install seo-skills@seranking

# Trigger with /seo-content-brief, /seo-page, etc.
# Update later with:
/plugin marketplace update seranking

Method 3: Manual install (only for seo-google or optional extensions)

Methods 1 and 2 cover 25 of the 26 skills. The 26th, seo-google, ships Python scripts that need a local clone. The optional Firecrawl and Google extensions also need this. Skip this method entirely if neither matters to you.

git clone --depth 1 https://github.com/seranking/seo-skills.git
bash seo-skills/install.sh

# Or as a one-liner:
curl -fsSL https://raw.githubusercontent.com/seranking/seo-skills/main/install.sh | bash -s -- --all

--all enables both extensions. Use --firecrawl or --google individually, or --no-extensions to clone only. Requires git, python3 3.10+, and node 20+ for Firecrawl. The manual install only wires your environment; you still need Method 1 or 2 so Claude can find the skill files.

Method 4: Cursor

Packaged as a Cursor plugin. Install from cursor.com/marketplace once listed, or add by name: /add-plugin se-ranking. The hosted SE Ranking MCP server auto-detects on install; open Cursor's MCP settings and sign in via OAuth, no API token to manage.

Connecting SE Ranking (any install path)

The SE Ranking remote MCP auto-registers when you install the plugin. On your first session, run /mcp and sign in via OAuth, no API token needed. An SE Ranking account with API access is required (sign up at seranking.com/api.html).

Manual fallback if /mcp does not list se-ranking after install:

claude mcp add --transport http se-ranking https://api.seranking.com/mcp
# Then run /mcp and complete OAuth

The 7 skills the README itself flags as the ones to reach for first, each with a real, dated public example run.

The 7 featured skills

seo-ai-search-share-of-voice: share-of-voice heatmap across ChatGPT, Perplexity, Gemini, Google AI Overview, and AI Mode, plus topic-cluster ownership per brand. Real example: 36.1% combined share of voice for wix.com across 5 LLM engines.

seo-backlink-gap: prospect list of referring domains linking to multiple competitors but not to your site, enriched with authority, anchor samples, and outreach angles.

seo-content-brief: writer-ready editor brief from a domain and topic, with keyword research, SERP analysis, competitor content teardown, internal linking plan, and an AI Search angle.

seo-keyword-cluster: intent-grouped clusters with a pillar-plus-spokes architecture and H1/H2 suggestions per spoke.

seo-ads: paid-search competitive intelligence. Domain ad footprint, bidding landscape per keyword, ad-copy patterns, SERP shopping/ad-pack visibility, recommended bid-keyword shortlist.

seo-content-audit: E-E-A-T (60-item) plus CITE (30-item) audit for existing content with a publish / publish-with-fixes / no-publish verdict and citation-readiness analysis.

seo-sxo: reads SERPs backwards to find page-type mismatches, scores the page from 4 personas, and produces a wireframe for the SERP-winning page type.

A real example worth knowing: the seo-plan composition

The repo's own flagship example (seo-plan for wix.com, a 90-day plan) is not a single skill's output. It composes results from seo-ai-search-share-of-voice, seo-competitor-gap-analysis, and seo-drift into one phased plan with owners, success metrics, and a critical path. This is the clearest demonstration in the whole repo of skills chaining into something bigger than any one of them alone.

The remaining 19 skills, covering everything from competitor gaps to a full API integration architect.

The 19 skills

seo-competitor-gap-analysis: keywords competitors rank for that the target does not, filtered by intent, volume, KD, and scored for priority.

seo-agency-landing-page: demand-gen landing page for an SEO agency with a free-audit lead magnet, grounded in real niche data.

seo-page: URL-level keyword and traffic intelligence with a keep / refresh / consolidate / kill verdict for one page.

seo-schema: JSON-LD detect, validate, generate. Five bundled templates (Article, Product, LocalBusiness, FAQPage, BreadcrumbList).

seo-images: image SEO audit for a URL or sampled domain: alt-text quality, WebP/AVIF coverage, srcset/sizes, lazy-loading and LCP signals, CLS dimensions, file naming, and ImageObject JSON-LD. Returns paste-ready snippets.

seo-drift: "git for SEO", baseline, compare, history. Severity-coded regression report across authority, traffic, keywords, backlinks, and page fingerprint.

seo-technical-audit: one-shot technical audit covering crawlability, indexability, security, mobile, structured data, and JS rendering, with a top-10 fix list ranked by impact times effort.

seo-sitemap: compares a domain's XML sitemap to the most recent SE Ranking audit, surfacing missing-from-sitemap pages, orphans, broken entries, and lastmod issues.

seo-competitor-pages: generates "X vs Y", "alternatives to X", and "best X for Y" landing pages with a feature matrix, schema, balanced verdict, and CTA flow.

seo-backlinks-profile: full backlink profile audit (versus gap-only seo-backlink-gap): authority distribution, anchor diversity, IP/subnet concentration, growth/decay, and toxic-candidate flagging.

seo-subdomain: subdomain ownership map, listing subdomains, topic ownership, fragmentation/cannibalisation flags, and consolidate/split recommendations.

seo-geo: URL-level Generative Engine Optimization analysis. AIO citation footprint per primary keyword, page passage-level audit, and recommendations to improve LLM citability.

seo-keyword-niche: mines longtail and question keywords for niche content opportunities, outputting a content-tier plan with template spec, URL pattern, sample pages, and thin-content quality gates.

seo-hreflang: hreflang and international SEO audit for multi-language/multi-region sites. Validates language-region codes, return tags, x-default, and canonical alignment. PASS/NEEDS-FIX/BROKEN verdict.

seo-local: local SEO audit for brick-and-mortar, service-area, and multi-location businesses. Scores Google Business Profile signals, NAP consistency, on-page local intent, citation footprint, and local-pack rank.

seo-firecrawl: ad-hoc web scraping, site mapping, full-site crawling, and within-domain search via Firecrawl. Returns raw HTML, JSON-LD, and screenshots WebFetch cannot. Requires the Firecrawl extension.

seo-plan: phased SEO roadmap for a domain, quarter by quarter, tied to competitive position, content gaps, technical debt, and AI Search readiness. Sequences specialist-skill outputs into one site-level plan.

seo-google: Google's own SEO data end to end (GSC, PageSpeed Insights, CrUX, Indexing API, GA4, YouTube, NLP, Knowledge Graph, Web Risk, Keyword Planner) across 4 credential tiers. Requires the Google APIs extension. Adapted with attribution from AgriciDaniel/claude-seo's upstream skill.

seo-api: SE Ranking API integration architect covering the entire 195-tool surface (Data API plus Project API), forecasting credit cost, emitting ready-to-paste cURL/Python/TypeScript/MCP recipes.

Two opt-in extensions. Skills that use them degrade gracefully when absent rather than failing.

Firecrawl (raw HTML, JSON-LD, JS rendering, site crawl)

Recommended but not mandatory for any skill inspecting raw HTML. Without it, those skills fall back to Claude's built-in WebFetch, which returns markdown only, stripping every meta tag, canonical link, and JSON-LD block before the skill sees it. Firecrawl closes that gap and adds full-site crawl, URL discovery, in-site search, and screenshots.

# From inside a clone of the repo
bash extensions/firecrawl/install.sh

Free tier is 500 credits/month at firecrawl.dev. The script prompts for your FIRECRAWL_API_KEY and merges the MCP entry into ~/.claude/settings.json without overwriting existing entries.

Skills that gain capability with Firecrawl: seo-page, seo-schema, seo-geo, seo-technical-audit, seo-sitemap, seo-content-audit, seo-drift, seo-content-brief, seo-competitor-pages, seo-sxo, seo-backlinks-profile. A --no-firecrawl flag opts out at runtime even when installed, to save credits.

Two skills are hard exceptions that need Firecrawl to run at all: seo-firecrawl itself, and seo-images, since srcset, loading, and fetchpriority attributes are stripped from WebFetch's markdown entirely.

Google APIs (real CWV, GSC, GA4, YouTube, Keyword Planner)

Needed for the seo-google skill specifically. All Google APIs used here are free with quotas. Setup is one-time, adapted with attribution from AgriciDaniel/claude-seo's upstream seo-google skill (MIT), same Python scripts namespaced to ~/.config/seo-skills/.

bash extensions/google/install.sh

4 credential tiers, each unlocking more: Tier 0 (API key) gives PSI, CrUX, YouTube, NLP, Knowledge Graph, Web Risk. Tier 1 (+service account) adds GSC and Indexing. Tier 2 (+GA4 property ID) adds GA4. Tier 3 (+Ads dev token) adds Keyword Planner.

Skills that gain capability when configured: seo-google itself (fully gated on this extension), plus seo-page, seo-drift, seo-technical-audit, seo-content-audit, seo-sitemap, seo-geo, seo-keyword-cluster, seo-keyword-niche, and seo-plan (real impressions/clicks/conversions/CWV instead of estimates).

All 26 skills, the agency workflow chain, rate limits, and full install commands.

The documented agency workflow chain

From the README, verbatim sequence:

1. Run seo-competitor-gap-analysis to see what opportunities exist.
2. Run seo-keyword-cluster on the gap output to plan content.
3. Run seo-content-brief on each cluster's pillar topic.
4. Run seo-backlink-gap to identify link targets for the new content.
5. Run seo-ai-search-share-of-voice quarterly to track LLM visibility.

Rate limits and cost management

The SE Ranking MCP has two namespaces: Data API (10 requests/second, charges credits per call) and Project API (5 requests/second, retrieval is free, writes consume quota). Every skill paces sequentially inside these limits.

The largest skills (seo-competitor-gap-analysis with full keyword dumps, seo-keyword-cluster with 20 seeds) can consume thousands of credits on large domains. Check DATA_getCreditBalance before running on production accounts, and use the ceiling parameters the skills document to cap cost.

Full 26-skill table

Skill	What it produces
seo-ai-search-share-of-voice	Share-of-voice heatmap across 5 LLM engines, topic ownership per brand
seo-backlink-gap	Referring domains linking to competitors but not you, enriched for outreach
seo-content-brief	Writer-ready brief: keywords, SERP, competitor teardown, internal linking
seo-keyword-cluster	Intent-grouped clusters, pillar plus spokes, H1/H2 suggestions
seo-ads	Paid-search competitive intelligence, ad footprint, bid shortlist
seo-content-audit	E-E-A-T + CITE audit with publish verdict
seo-sxo	SERP-backwards page-type analysis, 4-persona scoring
seo-competitor-gap-analysis	Keywords competitors rank for that you do not
seo-agency-landing-page	Demand-gen page with free-audit lead magnet
seo-page	URL-level keep/refresh/consolidate/kill verdict
seo-schema	JSON-LD detect, validate, generate (5 templates)
seo-images	Full image SEO audit, paste-ready snippets
seo-drift	Baseline/compare/history regression report
seo-technical-audit	Crawl/index/security/mobile audit, top-10 fix list
seo-sitemap	Sitemap vs audit comparison, orphans, broken entries
seo-competitor-pages	X vs Y / alternatives / best-X-for-Y pages
seo-backlinks-profile	Full backlink profile, toxic-candidate flagging
seo-subdomain	Subdomain ownership and cannibalisation map
seo-geo	URL-level GEO/AIO citation analysis
seo-keyword-niche	Longtail/question keyword content-tier plan
seo-hreflang	International SEO / hreflang audit
seo-local	Local SEO, GBP, NAP, citation audit
seo-firecrawl	Ad-hoc scraping/crawling orchestrator
seo-plan	Phased site-level SEO roadmap
seo-google	Full Google SEO data suite, 4 credential tiers
seo-api	SE Ranking API integration architect, 195-tool surface

Install everything

# Claude Code, fastest path
/plugin marketplace add seranking/seo-skills
/plugin install seo-skills@seranking

# For seo-google or the optional extensions, also run:
git clone --depth 1 https://github.com/seranking/seo-skills.git
curl -fsSL https://raw.githubusercontent.com/seranking/seo-skills/main/install.sh | bash -s -- --all

# Connect SE Ranking (any install path)
/mcp
# Sign in via OAuth