Collapse Detection

SkillDocs & knowledge

Documentation collapse anti-patterns - detection rules, bad examples, and remediation strategies for type-mixing violations

Available today. Use it from your connected AI after setup.

Connect ahel once, and every AI you use reads what you have installed.

Then ask your AI: use the Collapse Detection skill

What this skill tells your AI

The instructions your AI receives, as published by nwave-ai/nwave in nWave/skills/nw-collapse-detection/SKILL.md and read by ahel’s review.

Documentation collapse = types merge inappropriately, serving no audience well.

Anti-Patterns

Tutorial Creep

Description: Tutorial explains "why" extensively | Detection: explanation >20% | Fix: extract to separate doc, link back

How-to Bloat

Description: How-to teaches basics first | Detection: teaching fundamentals before steps | Fix: link to tutorial, assume baseline

Reference Narrative

Description: Reference includes conversational prose | Detection: prose paragraphs in entries | Fix: move to explanation doc, keep factual

Explanation Task Drift

Description: Explanation ends with "do this" steps | Detection: step-by-step in explanation | Fix: move steps to how-to, link

Hybrid Horror

Description: Single doc tries all four types | Detection: 3+ quadrants in one doc | Fix: split with clear boundaries

Detection Rules

Flag collapse when:

  • 20% content from adjacent quadrant | Two user needs served simultaneously | Ambiguous user journey stage

  • "Why" in tutorials | Task steps in explanations | Teaching in how-tos | Narrative in references

Bad Examples for Calibration

Tutorial with Task Focus

# Getting Started
If you need to deploy to production, follow these steps...

Problem: Assumes user knows what "deploy to production" means. A tutorial should assume nothing.

How-to Teaching Basics

# How to Configure Authentication
First, let's understand what authentication is. Authentication is...

Problem: Should assume user knows what authentication is. Link to a tutorial instead.

Reference with Opinions

## login(username, password)
This is probably the most important function you'll use...

Problem: Reference should be factual, not opinionated. Remove "probably the most important" editorializing.

Explanation with Steps

# Why We Use Microservices
... therefore, you should: 1. Create a service, 2. Deploy it...

Problem: Steps belong in a how-to guide. The explanation should end with reasoning and link to the how-to for action.

Signals

GitHub stars
610
Forks
63
Last commit
Sep 2026
Advanced
Catalog kind
skill
Gateway key
nw-collapse-detection
Source
github.com/nwave-ai/nwave