Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Une commande directe contourne le prompt de vérification. Examinez la source avant de l'exécuter.
Map 50-200+ keywords into topical clusters grouped by search intent. Build a content roadmap for dominating a topic with hub-and-spoke architecture. Google rewards topical authority — this skill builds the strategic map that tells you exactly what content to create and in what order.
Stage
S3: Blog & SEO — This is the strategic planning layer FOR blog content. Before writing individual posts, you need a map of the entire keyword landscape organized into clusters.
When to Use
User wants to plan SEO content strategy for a niche
User asks about keyword research, clustering, or topical authority
User says "keyword", "SEO plan", "content roadmap", "topic cluster", "hub and spoke"
Before running affiliate-blog-builder — to know WHICH articles to write
After monopoly-niche-finder — to map the keyword universe for the winning niche
Input Schema
niche:string# REQUIRED — the topic to cluster# e.g., "AI video tools", "email marketing for SaaS"seed_keywords:string[]# OPTIONAL — starting keywords to expand from# Default: auto-generated from nichedepth:string# OPTIONAL — "quick" (50 keywords) | "standard" (100) | "deep" (200+)# Default: "standard"affiliate_products:string[]# OPTIONAL — products you promote (to prioritize commercial keywords)# Default: none
Chaining from S1 monopoly-niche-finder: Use monopoly_niche.intersection as the niche input.
Workflow
Step 1: Generate Seed Keywords
If not provided, generate 5-10 seed keywords from the niche:
Identify the hub page (broadest, highest-volume cluster)
Connect spoke pages (specific clusters that link back to hub)
Prioritize by: commercial intent first (revenue), then informational (traffic)
Estimate effort: number of articles needed, suggested publishing cadence
Step 6: Self-Validation
Clusters are based on actual search data, not guesses
Each cluster has a clear search intent (I, C, or T)
Hub-and-spoke structure is logical (hub is broad, spokes are specific)
Priority ordering makes business sense (revenue-driving content first)
Total content pieces are realistic for user's capacity
Output Schema
output_schema_version:"1.0.0"keyword_clusters:niche:stringtotal_keywords:numbertotal_clusters:numberhub:keyword:stringcluster_name:stringcontent_type:stringpriority:numberclusters:-name:stringintent:string# "informational" | "commercial" | "transactional"hub_keyword:stringkeywords:string[]content_type:string# "blog" | "comparison" | "review" | "tutorial" | "landing"priority:number# 1-5estimated_volume:stringcontent_roadmap:total_articles:numberpublishing_cadence:stringpriority_order:string[]# Cluster names in order to writetarget_keywords:string[]# Flat list of all keywords for chainingchain_metadata:skill_slug:"keyword-cluster-architect"stage:"blog"timestamp:stringsuggested_next:-"affiliate-blog-builder"-"content-moat-calculator"-"comparison-post-writer"-"landing-page-creator"
### Clusters by Priority
#### Priority 1: [Cluster Name] (Commercial Intent)
- **Hub keyword:** [keyword] — [volume]
- **Content type:** [comparison / review]
- **Keywords:** [list]
- **Article idea:** [specific title]
#### Priority 2: [Cluster Name] (Informational Intent)
[same structure]
[Continue for all clusters]
### Content Roadmap
| Week | Cluster | Article | Intent | Priority |
|---|---|---|---|---|
| 1 | [cluster] | [title] | C | 1 |
| 2 | [cluster] | [title] | C | 1 |
| 3 | [cluster] | [title] | I | 2 |
### Next Steps
- Run `content-moat-calculator` to estimate effort for topical authority
- Run `affiliate-blog-builder` for Priority 1 articles
- Run `comparison-post-writer` for commercial clusters
Error Handling
Niche too broad: "This niche is very broad. Let me narrow to a sub-niche for more actionable clusters. Or run monopoly-niche-finder first."
No search volume: "This niche may be too narrow for significant search traffic. Consider broadening slightly."
Too many keywords: Group aggressively into fewer clusters. Quality of clustering > quantity of keywords.
No commercial intent keywords: Flag as concern — hard to monetize through affiliate without commercial intent. Suggest adjacent niches.
Examples
Example 1: "Map keywords for AI video tools"
→ Seeds: "best AI video tools", "AI video generator", "HeyGen review". Expand to 100+ keywords. Cluster: "AI video reviews" (C), "how to make AI videos" (I), "AI video pricing" (T), "AI video vs traditional" (C). Hub: "Best AI Video Tools 2025".
Example 2: "Keyword strategy for my affiliate blog about email marketing"
→ Deep keyword research. Clusters: "email marketing platforms" (C), "email automation tutorials" (I), "email marketing pricing comparison" (T), "email deliverability guides" (I).
Example 3: "Plan my content roadmap" (after monopoly-niche-finder)
→ Pick up niche from chain. Map 100+ keywords in that intersection niche. Prioritize clusters by revenue potential.
Flywheel Connections
Feeds Into
affiliate-blog-builder (S3) — which articles to write and target keywords
comparison-post-writer (S3) — commercial clusters become comparison articles
