What problem does it solve? Keyword grouping by text similarity often produces content plans where pages compete with each other or miss how Google actually ranks topics. This Skill clusters keywords by real SERP overlap (shared top-10 results), then designs a hub-and-spoke content architecture with internal link matrices and an interactive cluster map. ## Core Features & Use Cases - SERP-Overlap Clustering: Expands a seed keyword into 30-50 variants, compares pairwise top-10 organic results via WebSearch or DataForSEO, and merges or separates keywords using overlap thresholds. - Hub-and-Spoke Architecture: Selects a pillar page, groups 2-5 spoke clusters, assigns intent-based templates and word count targets, and generates a JSON internal link matrix with cannibalization checks. - Execution & Visualization: Generates an interactive cluster-map.html, optionally executes content creation through claude-blog, produces content briefs otherwise, and scores results with a cluster scorecard. - Use Case: Given the seed keyword "email marketing software", produce a pillar page plan with 3 spoke clusters, a full internal link matrix, and an interactive map showing which posts are written versus planned. ## Quick Start Ask the assistant to run a topic cluster plan for your seed keyword, for example: create a content cluster plan for "email marketing software".
