ultrathink-protocol

What problem does it solve?

It prevents wasted time caused by trial-and-error debugging when a system remains broken and the true root cause is unknown.

Core Features & Use Cases

• Explicit THOUGHT/ACTION/OBSERVATION cycles: Forces hypothesis testing with direct evidence instead of narrative troubleshooting.
• Verified-cause gate: Blocks implementation until the cause is confirmed with evidence from the running system (e.g., logs, process state, ports, config reads).
• Diagnosis ladder and escalation: Guides investigation from clearly stating the symptom to tracing the data path and reading source code, then escalates to research only when the question becomes answerable.

Use case examples: fixing a service that still fails after multiple restart/config changes; diagnosing multi-layer failures across runtime/middleware/network boundaries; breaking out of a looping pattern where each “fix” changes nothing but feels progress-heavy.

Quick Start

Use the ultrathink-protocol skill when a production incident is still unresolved after multiple unsuccessful fix attempts and you need to stop guessing and verify the root cause with direct evidence.