nav-diagnose

Detect and diagnose quality degradation in human-AI collaboration conversations.

32|3|Updated Jan 21, 2026

Search Tags:#quality assurance #debugging #error detection #prompt engineering #ai collaboration #theory of mind

Ready to run

npx skills add https://github.com/dkyazzentwatwa/SuperNavigator --skill nav-diagnose

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: nav-diagnose
Source: https://github.com/dkyazzentwatwa/SuperNavigator/tree/main/skills/os-layer/context-memory/nav-diagnose
Command: npx skills add https://github.com/dkyazzentwatwa/SuperNavigator --skill nav-diagnose

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes scripts (resource) and references (resource) components.

What problem does it solve?

This Skill addresses drops in AI collaboration quality by detecting issues like repeated corrections, context confusion, or user frustration, and then re-anchoring the AI's understanding.

Core Features & Use Cases

Quality Monitoring: Automatically detects signs of degrading AI performance and communication breakdowns.
Root Cause Analysis: Identifies potential reasons for quality drops, such as misunderstanding instructions or context drift.
Re-anchoring: Prompts the user to confirm the AI's understanding and realigns the conversation.
Use Case: If an AI repeatedly makes the same mistake after being corrected, this skill will trigger, diagnose the issue, and ask clarifying questions to get back on track.

Quick Start

Use the nav-diagnose skill to check for quality drops in our recent conversation.

Frequently Asked Questions about nav-diagnose

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

FAQPage Schema

How do I detect AI collaboration quality degradation during a conversation?▼

To detect AI collaboration quality degradation, analyze conversation patterns for indicators like repeated corrections, hallucinations, context confusion, and user frustration. This diagnostic process identifies root causes and re-anchors the AI's understanding to restore effective communication.

Why does an AI keep making the same mistake after I correct it?▼

An AI repeats mistakes after correction due to context confusion or context drift, where the AI loses track of instructions. Root cause analysis diagnoses this quality drop, triggering clarifying questions to re-anchor the AI's understanding and restore effective communication.

What is the best way to diagnose hallucinations and context confusion in prompt engineering?▼

The best way to diagnose hallucinations and context confusion is by monitoring conversation patterns for severity levels that trigger diagnostic alerts. This identifies root causes of misunderstanding instructions and suggests corrective actions to re-anchor the AI.

How do I re-anchor an AI's understanding when user frustration occurs?▼

To re-anchor an AI's understanding when user frustration occurs, prompt the user to confirm the AI's understanding and ask clarifying questions. This realigns the conversation and resolves communication breakdowns caused by context drift.

Can I use automated error detection to monitor AI communication breakdowns?▼

Yes, you can use automated error detection to monitor AI communication breakdowns by utilizing predefined patterns and severity levels. This automatically detects signs of degrading performance and triggers diagnostic alerts for root cause analysis.