Free SEO, GEO and AEO agent skills (SKILL.md) for Claude Code, Codex and Cursor

Six free, MIT-licensed agent skills. Install all of them with npx skills add cygnixlabs-cmd/cygnix-seo-skills. Each is a short SKILL.md plus one standard-library Python script: no API keys and no paid SEO tool.

keyword-clustering-lite

Group a keyword list into topic clusters with a pillar keyword, majority search intent and suggested page type, so one page targets each cluster. Use when the user pastes keywords or a keyword CSV and asks to cluster, group, organise or map keywords to pages, or plan a content hub. Offline, no API keys.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill keyword-clustering-lite

Real output from skills/keyword-clustering-lite/scripts/cluster.py examples/keywords.txt on the bundled example file:

| # | pillar keyword | intent | page type | keywords | members |
|---|---|---|---|---|---|
| 1 | crm for small business | commercial | comparison, 'best X' list or review page | 3 | best crm for small business; small business crm software |
| 2 | crm vs erp | commercial | comparison, 'best X' list or review page | 2 | erp vs crm difference |
| 3 | crm pricing | transactional | product, pricing or service page | 2 | crm pricing comparison |
| 4 | free crm software | commercial | comparison, 'best X' list or review page | 2 | crm software near me |
| 5 | what is a crm | informational | guide, how-to or FAQ article | 1 |  |
| 6 | hubspot login | navigational | brand, login or support page | 1 |  |
| 7 | how to choose a crm | informational | guide, how-to or FAQ article | 1 |  |

12 keywords -> 7 clusters (threshold 0.5). Lexical clustering: check top clusters against the live SERP before merging pages.

Need it at scale or inside a workflow? Apify actor: Keyword Clustering Tool & Grouper + Content Brief Generator · full skill on Agensi: Keyword Clusters + Intent Map: No Paid SEO Tool Needed

search-intent-lite

Classify keywords by search intent (informational, commercial, transactional, navigational, plus a local flag) and suggest the page type that should target each one. Use when the user asks for search intent, keyword intent, what kind of page to build for a keyword, or why a page type does not rank. Offline, rule-based, no API keys.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill search-intent-lite

Real output from skills/search-intent-lite/scripts/intent.py examples/keywords.txt on the bundled example file:

| keyword | intent | confidence | local | page type |
|---|---|---|---|---|
| best crm for small business | commercial | high |  | comparison, 'best X' list or review page |
| crm for small business | unclear | low |  | check the live SERP before choosing |
| small business crm software | commercial | high |  | comparison, 'best X' list or review page |
| crm pricing | transactional | high |  | product, pricing or service page |
| crm pricing comparison | transactional | mixed |  | product, pricing or service page |
| how to choose a crm | informational | high |  | guide, how-to or FAQ article |
| what is a crm | informational | high |  | guide, how-to or FAQ article |
| crm vs erp | commercial | high |  | comparison, 'best X' list or review page |
| erp vs crm difference | commercial | high |  | comparison, 'best X' list or review page |
| free crm software | commercial | high |  | comparison, 'best X' list or review page |
| hubspot login | navigational | high |  | brand, login or support page |
| crm software near me | commercial | high | yes | local landing / service-area page |

Totals: commercial 6, transactional 2, informational 2, unclear 1, navigational 1

Need it at scale or inside a workflow? Apify actor: SERP Intent Mapper: Keyword Search Intent Classifier · full skill on Agensi: SERP Intent Mapper — Keyword → Page Type Match

keyword-cannibalization-lite

Find queries where two or more of your own pages split Google impressions, from a Google Search Console query+page export, and rank them HIGH/MEDIUM/LOW. Use when the user asks about keyword cannibalization, competing pages, pages fighting for the same keyword, or uploads a GSC export with query and page columns. Offline, no API keys.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill keyword-cannibalization-lite

Real output from skills/keyword-cannibalization-lite/scripts/cannibalization.py examples/gsc-query-page.csv on the bundled example file:

| severity | query | primary page (most clicks) | competing pages | clicks on competitors | #2 impression share | position gap |
|---|---|---|---|---|---|---|
| HIGH | how to clean running shoes | https://example.com/blog/clean-shoes | https://example.com/blog/shoe-care | 22 | 49% | 4.4 |
| HIGH | best running shoes | https://example.com/best-running-shoes | https://example.com/blog/running-shoes-guide | 15 | 31% | 8.6 |

3 queries checked, 2 cannibalized (2 HIGH). Thresholds: >= 10 impressions and >= 10% share per page.

Need it at scale or inside a workflow? Apify actor: Keyword Cannibalization Checker for Google Search Console · full skill on Agensi: Fix Keyword Cannibalization Using Google Search Console

content-decay-lite

Compare two Google Search Console page exports (an earlier period and a recent one) and flag pages that are LOST, DECLINING, in STRIKING_DISTANCE, GROWING or STABLE, ranked by clicks lost. Use when the user asks which pages lost traffic, about content decay, what to refresh or update first, or uploads two GSC page exports. Offline, no API keys.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill content-decay-lite

Real output from skills/content-decay-lite/scripts/decay.py examples/gsc-pages-previous.csv examples/gsc-pages-current.csv on the bundled example file:

| page | status | clicks before | clicks now | click change | % change | position before → now |
|---|---|---|---|---|---|---|
| https://example.com/seo-guide | DECLINING | 800 | 420 | -380 | -48% | 4.2 → 6.1 |
| https://example.com/keyword-research | DECLINING | 210 | 90 | -120 | -57% | 8.9 → 9.4 |
| https://example.com/link-building | DECLINING | 120 | 35 | -85 | -71% | 9.0 → 14.2 |
| https://example.com/old-post | LOST | 45 | 0 | -45 | -100% |  |
| https://example.com/new-post | STRIKING_DISTANCE | 0 | 60 | +60 |  |  |
| https://example.com/technical-seo | GROWING | 300 | 380 | +80 | +27% | 4.0 → 3.6 |

6 pages compared. Clicks lost on declining pages: 630. DECLINING 3, GROWING 1, LOST 1, STRIKING_DISTANCE 1

Need it at scale or inside a workflow? Apify actor: Content Decay Audit - Google Search Console Traffic Drop · full skill on Agensi: Find Decaying Pages: GSC Triage + Content Refresh Plan

llms-txt-check

Check whether a website has a valid /llms.txt (llmstxt.org format, working links, size) and whether robots.txt lets AI crawlers such as GPTBot, OAI-SearchBot, ClaudeBot, PerplexityBot and Google-Extended in. Use when the user asks about llms.txt, AI crawler access, blocking or allowing AI bots, or whether ChatGPT/Claude/Perplexity can read their site. Read-only.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill llms-txt-check

Real output from skills/llms-txt-check/scripts/llms_check.py plausible.io on the live site on 23 October 2026:

# llms.txt + AI crawler check: https://plausible.io

## /llms.txt
- **Missing** (HTTP 404).
- /llms-full.txt: not found (optional)

## AI crawler access (robots.txt, site root)
- robots.txt: found; Sitemap line: yes
| crawler | access to / |
|---|---|
| GPTBot | allowed |
| OAI-SearchBot | allowed |
| ChatGPT-User | allowed |
| ClaudeBot | allowed |
| Claude-SearchBot | allowed |
| PerplexityBot | allowed |
| Google-Extended | allowed |
| CCBot | allowed |

## Verdict
- Publish `/llms.txt`: an H1 with your site name, a `>` one-line summary, then `##` sections listing your 10-20 most useful pages as `- [Title](url): one-line description`.
- Note: AI search crawlers rarely fetch llms.txt today; coding agents do. Crawler access in robots.txt matters more for AI search.

Need it at scale or inside a workflow? Apify actor: AEO & AI Visibility Audit: LLM SEO, GEO + llms.txt Checker · full skill on Agensi: Ship llms.txt + Unlock AI Crawlers: Access Fix Pack

ai-visibility-quick-score

Give one web page a quick 0-100 AI search visibility (GEO/AEO readiness) score from its server-rendered HTML - AI crawler access, Organization/WebSite JSON-LD, structured data, title/meta, question headings, answer-first opening, llms.txt and freshness - with the top 5 fixes. Use when the user asks if a page is ready for ChatGPT, Perplexity, Google AI Overviews or other AI search, or wants a GEO/AEO score.

npx skills add cygnixlabs-cmd/cygnix-seo-skills --skill ai-visibility-quick-score

Real output from skills/ai-visibility-quick-score/scripts/quick_score.py https://plausible.io/ on the live site on 23 October 2026:

# AI visibility quick score: https://plausible.io/

**72/100 (GOOD)**, one page, server HTML only.

| check | points |
|---|---|
| AI crawler access (robots.txt) | 25/25 |
| Organization / WebSite entity (JSON-LD) | 15/15 |
| Other structured data | 0/10 |
| Title + meta description | 7/10 |
| One H1 + question-style headings | 15/15 |
| Answer-first opening + server-rendered text | 10/15 |
| /llms.txt | 0/5 |
| Freshness date signal | 0/5 |

## Top fixes
1. Add JSON-LD that matches the page (Article, Product, FAQPage, BreadcrumbList…)
2. Open with a 15-80 word direct answer (first real paragraph is 12 words)
3. Publish /llms.txt listing your 10-20 key pages (cheap; coding agents read it)
4. Show a visible Updated date and add dateModified in JSON-LD
5. Title should be 15-65 chars (now 75)

AI crawlers: GPTBot ok, OAI-SearchBot ok, ChatGPT-User ok, ClaudeBot ok, Claude-SearchBot ok, PerplexityBot ok, Google-Extended ok, CCBot ok
JSON-LD types: WebSite

This measures readiness (can AI crawlers fetch, parse and trust this page), not rankings or citations.

Need it at scale or inside a workflow? Apify actor: Website SEO Audit & Site Audit Crawler: Technical SEO + AEO/GEO · full skill on Agensi: AI Search Visibility Suite: 0-100 GEO Score + Report

Paid SEO and AI-search skills on Agensi

The full versions (more checks, report templates, fix plans) are sold on Agensi and are searchable from Claude and ChatGPT through the Agensi MCP connector.

• SEO Content Brief — Writer-Ready Outline: Free SEO content brief generator: give your agent a keyword and get a writer-ready brief and blog outline: search intent, H2/H3 article outline, questions to answer, internal links and an on-page SEO checklist.
• Find Decaying Pages: GSC Triage + Content Refresh Plan: Content decay audit for Google Search Console: find pages losing traffic, label each drop (DECAYING, RANK_SLIP, CTR_DROP, LOST, STRIKING_DISTANCE), rank by clicks lost and get a content refresh plan.
• SERP Intent Mapper — Keyword → Page Type Match: Search intent classifier + SERP check: label each keyword's intent, count the top 10 by page type (e.g. 9/10 comparisons) and flag your pages built in the wrong format: MATCH, PARTIAL or MISMATCH.
• Keyword Clusters + Intent Map: No Paid SEO Tool Needed: Keyword clustering + search intent map: group keywords into topic clusters by intent, pick pillar pages and a priority shortlist. Paste an Ahrefs/Semrush/GSC export for real numbers; zero invented search volumes.
• Citation Freshness Audit: dateModified + Sitemap: Unlike content-decay (GSC traffic) or schema type audits — checks visible Updated dates, JSON-LD dateModified, and sitemap lastmod for AI-citation freshness.
• Map Niche into Pillars Clusters and Internal Links: Unlike free topical-map or keyword-cluster prompts: a script clusters your real keywords into pillar + cluster pages, link anchors and a publish order.
• Search Console Quick Wins: Striking-Distance + CTR Fixes: Unlike cannibalization/decay audits: ranks GSC striking-distance (pos 4–20) + low-CTR pages by est. click upside, drafts titles/metas from real queries.
• Find Orphan Pages and Fix Your Internal Linking: Internal link audit from a crawl export: find orphan pages, underlinked money pages and weak anchors, then get exact source → target internal links to add, with anchor text, to fix site architecture.
• Win Featured Snippets: Ownership Map + PAA Bank: Map featured-snippet ownership and People Also Ask questions from real SERP checks, then score TARGET opportunities and answer drafts before you write.
• Get More YouTube Clicks: Keyword, Title + Chapter Audit: Map video keywords, score titles for click-through structure, audit chapters/descriptions/tags, find competitor topic gaps and run a publish checklist with ranked fixes. 8-tab workbook + 35 prompts.
• Win Featured Snippets + People Also Ask: SERP Win Plan: Map who owns the snippet for each query, bank real PAA questions, draft answers in the winning format, tear down competitor snippets and get a DO NOW fix list. 8-tab workbook + ~40 prompts.
• Plan Programmatic SEO Pages Without Thin Content: Turn a dataset plus one template into a pSEO page plan (URL, title, H1, meta), and gate every row PUBLISH / FIX / HOLD for thin data, near-duplicates, length and demand before you build.
• Local AI Visibility Audit: Does ChatGPT Recommend You?: Local SEO + AI visibility audit: see if ChatGPT, Perplexity and Google AI name your business for "best [service] in [city]", who gets named instead, and the GBP, NAP and city-page fixes.
• Hreflang Checker & Fixer: Return Tags, Codes, x-default: Audit hreflang on live pages: return tags, self tags, ISO codes (en-UK, jp, es-419), x-default, redirects, noindex and canonicals. Root-cause report + the corrected tag set, ready to paste.
• AEO Audit — Get Cited in ChatGPT + AI Overviews: For SEOs and content teams: check if ChatGPT, Perplexity and Google AI Overviews can cite your page, get answer-first rewrites and sample FAQ answers to win AI citations.
• Ship llms.txt + Unlock AI Crawlers: Access Fix Pack: llms.txt generator + AI crawler access fix: writes a starter llms.txt, audits robots.txt for GPTBot, ClaudeBot, PerplexityBot and Google-Extended, and ships a fix checklist so AI search can read your site.
• Fix Keyword Cannibalization Using Google Search Console: Keyword cannibalization checker for Google Search Console: find competing pages that split clicks for the same query, scored HIGH/MED/LOW, with a merge, 301, canonical or differentiate fix per page.
• Social Share Preview Audit: OG + Twitter Cards: Unlike meta-tag generators — offline QA of og:* + twitter:* preview readiness (missing tags, bad image URL, title truncation, title drift) before you post.
• AI Search Visibility Suite: 0-100 GEO Score + Report: GEO audit + AEO audit in one: score any site 0-100 for AI search visibility in ChatGPT, Perplexity, Claude and Google AI Overviews. Checks llms.txt, AI crawler access, schema, freshness. Client report + fix plan.
• Local SEO Keyword Pack — Service × City Matrix + Page Plan: Local keywords for any service business: service × city × "near me"/cost/urgent matrix, Google Business Profile categories and services, and a page plan with no doorway pages.
• Migrate Sites Without Losing SEO: Redirect + Parity Audit: Build an old→new redirect map, catch chains and loops, check title/H1/canonical/hreflang/schema parity, run pre/post-launch QA and flag traffic drops. Includes 8-tab workbook (blank + example) and 35 prompts. No made-up data.
• Local SEO Audit: GBP + NAP Fix Plan for Small Business: Finds the NAP errors that cost local businesses calls: click-to-call links that dial the wrong number, emails that go to the wrong inbox, schema with no address, then hands you a P0–P2 fix plan. Evidence only, no invented Maps ranks.
• SEO Meta Tags Generator — Titles, Descriptions + Schema: Unlike free keyword-only meta generators: 3–5 titles + 2–3 descriptions with char counts, OG/Twitter tags, evidence-only JSON-LD, from your real page.
• Keyword Gap Analysis — Competitor Gaps + Priority Plan: Keyword gap analysis from your Ahrefs, Semrush or GSC exports: label every keyword MISSING, BEHIND, WEAK or UNIQUE vs up to 5 competitors, then get a ranked page plan. No invented volumes.

Updated 23 October 2026. Built by Cygnix Labs. Prices and free tiers are read from the public Apify Store API on the date above; check the listing for the current price. llms.txt