Isolate failing infrastructure layers using evidence-based diagnostic checks.
npx skills add https://github.com/iuliandita/skills --skill debug-triage-iuliandita
Please help me install this Agent Skill. Skill: debug-triage Source: https://github.com/iuliandita/skills/tree/main/skills/debug-triage Command: npx skills add https://github.com/iuliandita/skills --skill debug-triage-iuliandita
This skill solves the problem of "incident noise" where engineers waste time guessing the cause of a system failure. It provides a structured, evidence-based method to isolate the failing layer in a complex stack before any diagnostic or remediation work begins.
Use the debug-triage skill to analyze the current incident and identify the failing layer based on the provided symptoms.
High-intent search queries and answers about installing and using this skill.
To troubleshoot an unknown production incident, systematically isolate failing infrastructure layers by running evidence-based diagnostic checks. This method excludes healthy layers like DNS and TLS first, preventing wasted time on unnecessary deep dives into the wrong component.
Triage Kubernetes 502 errors by applying layer localization to determine if a misconfigured ingress, dead backend pod, or network routing failure is responsible. This process uses non-destructive signal analysis to isolate the exact failing component before deeper investigation.
Yes, you need Kubernetes CLI access and standard Linux diagnostic utilities to perform system outage triage. These tools execute the non-destructive signal analysis required to systematically investigate cross-layer failures during production incidents.
The best way to isolate live system incidents is to use the cheapest discriminating checks first, systematically excluding healthy layers like DNS, Network, and Ingress. This evidence-based approach localizes the failure before routing to a specific domain skill.
After triage localizes the failing infrastructure layer, the process triggers an explicit handoff. It routes the incident directly to the specific domain skill responsible for that layer, such as networking, databases, or kubernetes, for targeted remediation.
You should avoid this systematic triage approach when the failing infrastructure layer is already known, or when destructive diagnostic methods are required. It is specifically designed for unknown root causes needing cross-layer investigation using non-destructive checks.
alex-voloshin-devDiagnose local Docker environments by analyzing container status, logs, networking, and resource utilization.
avav25Collect read-only evidence from Kubernetes clusters, databases, logs, and networking to diagnose root causes.
AnastasiyaWAudit Prometheus alert rules using the RED method.
ROYCE-8425Guide Clean Architecture implementation for Android and Kotlin Multiplatform projects.
