investigate-root-cause

Orchestrate four-agent ensemble investigations to determine health condition root causes.

Updated Jan 19, 2026
One-click install
npx skills add https://github.com/tsilva/health-agent --skill investigate-root-cause
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: investigate-root-cause
Source: https://github.com/tsilva/health-agent/tree/main/.claude/skills/health-agent/investigate-root-cause
Command: npx skills add https://github.com/tsilva/health-agent --skill investigate-root-cause

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This Skill provides a structured, multi-agent ensemble approach to identifying the root causes of health conditions. By combining four distinct reasoning strategies, it reduces cognitive bias, promotes diagnostic diversity, and delivers a calibrated confidence assessment with actionable follow-up.

Core Features & Use Cases

  • Four parallel agents with different reasoning strategies (Bottom-Up, Top-Down, Genetics-First, Red Team) for comprehensive coverage.
  • Mandatory evidence verification and cross-agent refinement to enhance reliability and minimize bias.
  • Adversarial validation and counter-hypothesis generation to surface alternative explanations.
  • Integrated data sources: labs, medical timeline, current medications, genetics, and literature.
  • Per-hypothesis falsification criteria, confidence calibration, and epidemiological priors where applicable.
  • Health state update with prioritized follow-up actions, goals, and diagnostic gaps.

Quick Start

  1. Initiate an ensemble investigation for a target condition (e.g., "investigate root cause of chronic fatigue").
  2. The system will spawn four parallel agents to generate initial findings.
  3. After completion, refinement and verification phases will produce a calibrated consensus report.
  4. The health state is updated with hypotheses, gaps, and suggested next steps.

Frequently Asked Questions about investigate-root-cause

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

FAQPage Schema
How do I investigate the root cause of a health condition when standard assessments are inconclusive?

To investigate the root cause of an inconclusive health condition, this Skill orchestrates a four-agent ensemble combining bottom-up, top-down, genetics-first, and red team reasoning. It integrates labs, medical timelines, and literature to yield a calibrated confidence report.

What is differential diagnosis and how does multi-agent ensemble validation work?

Differential diagnosis through multi-agent ensemble validation works by spawning four parallel agents with distinct reasoning strategies. It applies adversarial validation and counter-hypothesis generation to surface alternative explanations and minimize cognitive bias.

How do I use genetics and health data to verify root-cause hypotheses?

To use genetics and health data for root-cause hypothesis verification, the system applies mandatory evidence verification and per-hypothesis falsification criteria. It cross-refines findings from multiple agents against integrated data sources and literature.

Can I perform a root-cause investigation for complex cases without external dependencies?

Yes, you can perform a root-cause investigation for complex cases without external dependencies. The Skill operates autonomously, spawning parallel agents that integrate available medical timelines, current medications, and genetics to generate actionable follow-up steps.

What is the best way to calibrate confidence for a differential diagnosis?

The best way to calibrate confidence for a differential diagnosis is through cross-agent refinement and epidemiological priors. The ensemble generates falsification criteria for each hypothesis, producing a calibrated consensus report with prioritized follow-up actions.