What problem does it solve?
Network troubleshooting often devolves into guess-and-swap: engineers pattern-match to familiar causes, swap components, and fix symptoms instead of root causes. This Skill enforces the Kepner-Tregoe rational-process method so an agent specifies the fault boundary before hypothesizing, tests candidate causes against evidence, verifies the root cause before changing anything, and separates the fast reversible incident fix from the permanent fix.
Core Features & Use Cases
- Problem Analysis: Build an IS/IS-NOT specification across WHAT, WHERE, WHEN, and EXTENT dimensions, generate causes only from observed distinctions and changes, and verify the most probable cause with a read-only test before acting.
- Situation Appraisal: Triage alert storms by rating concerns on Seriousness, Urgency, and Growth, collapsing correlated alarms into one root, and hunting the outlier that signals a second incident.
- Decision Analysis & Potential Problem Analysis: Choose fixes with MUST/WANT screening plus adverse-consequence review, and protect planned changes with preventive and contingent actions that have explicit triggers and owners.
- Use Case: During an outage where checkout, payments, and a warehouse app all fail while a core switch logs errors, use the Skill to collapse the symptoms to one infrastructure fault, specify the boundary, verify a misconfigured LAG change as the cause, apply a reversible rollback, and protect the permanent fix.
Quick Start
Use the kepner-tregoe-network-troubleshooting skill to diagnose why users at one branch site cannot reach the ERP application while all other sites work fine.
