cannibalization-finder

Detects same-domain pages competing for identical search intent and ranks cannibalization clusters by severity.

Updated May 23, 2026
One-click install
npx skills add https://github.com/TheGraphicKing/skott-canvas --skill cannibalization-finder-thegraphicking
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: cannibalization-finder
Source: https://github.com/TheGraphicKing/skott-canvas/tree/main/skills/cannibalization-finder
Command: npx skills add https://github.com/TheGraphicKing/skott-canvas --skill cannibalization-finder-thegraphicking

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? When multiple pages on the same domain target the same search intent, they split ranking signals and suppress each other in search results. This Skill identifies those competing pages by clustering them on intent overlap rather than surface keyword matching, so you can consolidate, differentiate, or redirect before rankings plateau. ## Core Features & Use Cases - Intent-Based Clustering: Classifies pages across intent type, topic cluster, specificity level, and query form to compute a 0-1 intent overlap score, catching cannibalization even between pages using different keyword variants. - Severity Scoring: Applies an organic footprint filter using estimated traffic and inbound links, then classifies clusters as low, moderate, or critical severity. - Internal Link Exclusions: Generates URL pairs that must not receive mutual internal links until cannibalization is resolved, preventing link-building workflows from reinforcing the problem. - Use Case: A page's ranking has plateaued despite strong content and backlinks. Run this Skill against the domain to discover that two blog posts answer the same informational query, then consolidate the weaker page into the canonical one with a 301 redirect. ## Quick Start Analyze the domain example.com for pages competing with https://example.com/blog/on-page-seo-guide targeting the keyword on-page SEO, and return ranked cannibalization clusters with severity scores.

Frequently Asked Questions about cannibalization-finder

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I find keyword cannibalization on my website?▼

Keyword cannibalization detection works by comparing each same-domain page's intent classification against a target page across intent type, topic cluster, specificity, and query form. Pages scoring above the similarity threshold (default 0.65) are grouped into cannibalization clusters with severity ratings.

How to detect cannibalization between pages using different keywords?▼

Intent-based clustering catches cannibalization even when pages use different keyword variants, because it compares the underlying query each page answers rather than surface keyword matching. Two pages answering the same question with different phrasing still score high on intent overlap.

What severity levels does cannibalization analysis produce?▼

Clusters are classified as low, moderate, or critical based on intent overlap score: 0.65-0.79 is low, 0.80-0.89 is moderate, and 0.90 or above is critical. Pages with zero traffic and zero inbound links are downgraded to low severity regardless of score.

Can I adjust the intent similarity threshold for cannibalization detection?▼

Yes, the intent_similarity_threshold parameter accepts a value between 0 and 1, defaulting to 0.65. Raising it narrows detection to only near-identical intent matches, while lowering it surfaces broader topical overlap between pages.

What are the limitations of intent-based cannibalization detection?▼

The analysis depends on the quality of the domain page corpus and estimated traffic data; pages with missing metadata may be misclassified. It also evaluates intent overlap rather than actual SERP behavior, so confirmed ranking competition should be verified against live search results.