What do the different severity tiers mean?

Severity is based on the similarity score: over 90% is critical (near-duplicates), 70-90% is high, and 50-70% is medium overlap. Suggestions are prioritized based on these tiers.

Can it distinguish between a tutorial and a technical reference?

Yes, the skill is context-aware. It respects document purposes (SPEC, REF, HOWTO, RUNBOOK) and will suggest keeping both files if they serve different audience needs, even if they share similar topics.

How does it determine if two documents are similar?

It uses a multi-pass approach: computing the Jaccard index on normalized heading structures, extracting topic keywords from metadata, and performing content hashing to find exact paragraph matches.

Does this skill automatically delete or merge my files?

No, the Redundancy skill is strictly read-only. It provides a detailed report and suggestions for consolidation but never modifies, merges, or deletes your documentation files.

DocOps Redundancy Checker

Name: DocOps Redundancy Checker
Author: murillodutt

bymurillodutt

0•

コンテンツ管理

Detects and reports redundant or duplicate content across documentation files to optimize information architecture.

The DocOps Redundancy skill streamlines documentation maintenance by identifying structural overlaps and exact duplicates within a project's documentation tree. By analyzing heading structures with Jaccard similarity, extracting topic keywords, and performing content hashing, it classifies redundancies into severity tiers and provides intelligent consolidation strategies. Designed for high-scale documentation projects, this skill helps technical writers and developers maintain a single source of truth without risking accidental data loss through its strictly read-only reporting approach.

主な機能

01Context-aware consolidation suggestions tailored to document purpose (SPEC/REF/HOWTO)

020 GitHub stars

03Categorized severity reporting from critical to low-level overlap

04Automated topic keyword extraction from titles, frontmatter, and metadata

05Structural similarity analysis using Jaccard index on heading hierarchies

06Exact duplicate detection via content hashing of paragraphs and sections

ユースケース

01Consolidating overlapping technical specifications during project migrations

02Validating documentation structure after major product updates to ensure clarity

03Auditing large documentation repositories to identify and eliminate content bloat

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add murillodutt/cellm redundancy

For use in Claude.ai and ChatGPT