skill-overlap

Detects redundant SKILL.md pairs in a skill pack via cosine similarity analysis.

30|12|Updated Jun 21, 2026
One-click install
npx skills add https://github.com/anthony-chaudhary/fak --skill skill-overlap
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: skill-overlap
Source: https://github.com/anthony-chaudhary/fak/tree/main/.claude/skills/skill-overlap
Command: npx skills add https://github.com/anthony-chaudhary/fak --skill skill-overlap

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) components.

What problem does it solve?

This skill addresses the silent bloat in a skill pack by identifying pairs of SKILL.md files that contain highly redundant content, preventing maintenance overhead and discoverability issues.

Core Features & Use Cases

  • Pairwise Similarity Analysis: Uses cosine similarity on token frequency to detect overlapping content between skills.
  • Proposal-Only Workflow: Flags candidates for human review without performing any destructive actions like merging or deleting files.
  • Use Case: When a repository grows to dozens of skills, use this to find two skills that describe nearly identical functionality so they can be consolidated into a single, authoritative source.

Quick Start

Run the skill overlap detector to scan the current skills directory and print any redundant pairs found above the default similarity threshold.

Frequently Asked Questions about skill-overlap

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

FAQPage Schema
How do I find redundant skill documentation in a repository?

Detect redundant documentation by running a pairwise cosine similarity analysis on tokenized body text, which flags highly similar SKILL.md files for review and repository consolidation.

Does the redundancy analysis modify or delete duplicate SKILL.md files?

No, the redundancy analysis operates strictly as a non-destructive advisory tool that flags duplicate SKILL.md candidates for human-led refactoring without performing any merging or file deletion.

Can I detect overlapping skill content without installing external dependencies?

Yes, you can detect overlapping content without external dependencies because the similarity analysis runs directly on token frequency from the text body, requiring no additional packages or libraries.

How do I identify duplicate SKILL.md files for repository cleanup?

Identify duplicate SKILL.md files for repository cleanup by scanning the directory and evaluating pairwise text similarity, producing a list of redundant pairs that exceed the similarity threshold.

Are frontmatter headers included in the text similarity analysis?

No, frontmatter headers are excluded from the text similarity analysis, which specifically tokenizes the body text to ensure redundancy detection focuses purely on actual content overlap.