What does the Introspect tool do?

The Introspect tool asks each LLM surgeon to report their current capabilities and limitations, helping you understand what they can and cannot do effectively.

How does the Challenge feature improve code quality?

It provides a counter-argument to your proposed approach, highlighting potential architectural blind spots or security risks before you commit to implementation.

Does this skill require local models?

While part of the 3-surgeons ecosystem, the Neurologist specifically utilizes local models like Qwen3-4B to perform introspection and health checks efficiently.

When should I run a Neurologist Pulse?

Run a pulse at the start of a session, after infrastructure changes like model swaps, or if you notice slow or unexpected responses from the system.

What is the primary role of the Neurologist skill?

It monitors the health of the 3-surgeons system, challenges proposed changes for logic gaps, and audits model capabilities to ensure high-reliability outputs.

Neurologist System Monitor & Skeptic

Name: Neurologist System Monitor & Skeptic
Author: supportersimulator

bysupportersimulator

0•

分析与监控

Monitors system health and challenges model assumptions to ensure reliability and corrigibility in multi-LLM workflows.

The Neurologist skill acts as the diagnostic and critical core of the 3-surgeons multi-model consensus system, providing essential health monitoring and logic verification. It offers 'Pulse' checks to ensure that LLM backends, state stores, and GPU locks are operational before critical tasks, preventing infrastructure-related failures. Beyond simple monitoring, it serves as a 'Corrigibility Skeptic' that proactively challenges architectural decisions and security assumptions to reveal hidden blind spots. By facilitating 'Introspection' reports from other surgeons, it allows developers to calibrate their confidence in model outputs and verify capabilities immediately after infrastructure changes or model swaps.

主要功能

01Corrigibility challenges to stress-test architectural and security assumptions

02Severity-graded feedback with suggested verification tests for logic gaps

03Real-time system health 'Pulse' monitoring for LLM backends and storage

04Automated GPU lock verification to prevent resource hanging

050 GitHub stars

06Self-introspection reporting to audit individual surgeon capabilities

使用场景

01Calibrating model confidence after updating endpoints or swapping LLM backends

02Stress-testing high-stakes architectural or security-sensitive code changes

03Verifying system readiness before executing critical multi-model operations

What are Skills?·How to Install

Install with 🐟 Skill.Fish

npx skillfish add supportersimulator/3-surgeons neurologist

For use in Claude.ai and ChatGPT

主要功能

01Corrigibility challenges to stress-test architectural and security assumptions

02Severity-graded feedback with suggested verification tests for logic gaps

03Real-time system health 'Pulse' monitoring for LLM backends and storage

04Automated GPU lock verification to prevent resource hanging

050 GitHub stars

06Self-introspection reporting to audit individual surgeon capabilities

使用场景

01Calibrating model confidence after updating endpoints or swapping LLM backends

02Stress-testing high-stakes architectural or security-sensitive code changes

03Verifying system readiness before executing critical multi-model operations