One-click install
npx skills add https://github.com/tylerburleigh/claude-model-chorus --skill thinkdeep
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: thinkdeep
Source: https://github.com/tylerburleigh/claude-model-chorus/tree/main/skills/thinkdeep
Command: npx skills add https://github.com/tylerburleigh/claude-model-chorus --skill thinkdeep

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires anthropic-cli, google-generativeai, openai-cli.

What problem does it solve?

This Skill transforms complex debugging and problem-solving into a systematic, multi-step investigation. It eliminates guesswork by tracking hypotheses, accumulating evidence, and building confidence, helping you pinpoint root causes faster and more reliably.

Core Features & Use Cases

  • Hypothesis Tracking: Form, test, and refine theories as evidence accumulates.
  • Confidence Progression: Watch your certainty grow from 'exploring' to 'certain' with each step.
  • State Persistence: Pause and resume complex investigations across sessions without losing context.
  • Use Case: Debug an intermittent 500 error in production. This Skill guides you through examining logs, code, and configurations, tracking your hypotheses and confidence until the root cause is identified.

Quick Start

Investigate why API latency increased from 100ms to 2s after deployment, starting with an analysis of recent changes.

Frequently Asked Questions about thinkdeep

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

FAQPage Schema
How do I systematically debug complex problems with hypothesis tracking?

Systematic debugging with hypothesis tracking lets you form, test, and refine theories as evidence accumulates. This Skill maintains your hypotheses and confidence levels across investigation steps, helping you pinpoint root causes faster than guesswork approaches.

Can I pause and resume debugging investigations across multiple sessions?

Yes, this Skill maintains state persistence across turns, allowing you to pause complex investigations and resume them later without losing your hypothesis history, accumulated evidence, or confidence progression.

What types of problems can I solve with structured hypothesis tracking?

Structured hypothesis tracking applies to debugging intermittent errors, root-cause analysis, security investigations, performance optimization, and architecture decisions. The Skill guides evidence accumulation and confidence evolution until the cause is identified.

How does confidence progression work during an investigation?

Confidence progression tracks your certainty level from 'exploring' through to 'certain' as you gather evidence. Each investigation step refines your hypotheses and updates your confidence, providing visibility into how close you are to identifying the root cause.

Can I use this for analyzing API latency spikes after deployment?

Yes, this Skill is designed for multi-step investigations like diagnosing API latency increases. It systematically guides you through examining recent changes, logs, code, and configurations while tracking hypotheses and building evidence toward root cause identification.