# Troubleshoot DevOps Incidents Systematically

Complex incidents often produce scattered signals across applications, infrastructure, and cloud services. This skill organizes evidence, tests hypotheses, and recommends immediate and preventive actions.

## Install

```bash
npx skillstore add sickn33/devops-troubleshooter
```

## Metadata

- Status: approved
- Slug: sickn33-devops-troubleshooter
- Skillstore revision: r2
- Version status: missing
- Tree hash: 0de486057a75c5e5cd6170614ffa09fde19f3f3092630eaa8f857ecf5fe71d46
- Author: sickn33
- GitHub username: sickn33
- License: MIT
- Repository: https://github.com/sickn33/antigravity-awesome-skills/tree/main/skills/devops-troubleshooter
- Ref: f9e2c34b4f19c7f3e6b0a1e93227b5f77cc12526
- Supported tools: Claude, Codex, Claude Code
- Audit status: complete
- Agent install advisory: allowed
- Manual install advisory: allowed
- Artifact signature: available
- Audit attestation: unavailable
- Human verification: not\_verified
- Risk factors: external\_commands
- Quality score: 79
- Quality tier: bronze
- Public page: https://skillstore.pages.dev/skills/sickn33-devops-troubleshooter
- Manifest: https://skillstore.pages.dev/api/skills/sickn33-devops-troubleshooter/manifest

## Capabilities

- Structures incident assessment around impact, scope, urgency, and available evidence.
- Guides analysis of logs, metrics, traces, health checks, and system state.
- Develops prioritized hypotheses and verification steps with minimal production impact.
- Covers Kubernetes, cloud platforms, networking, databases, CI/CD pipelines, and distributed services.
- Separates immediate service restoration from permanent remediation and resilience improvements.
- Produces postmortem inputs, monitoring recommendations, and reusable runbook guidance.

## Use Cases

- Triage a Production Outage: Organize symptoms, assess impact, and build a safe investigation sequence for restoring service.
- Diagnose Kubernetes Failures: Analyze pod restarts, resource pressure, networking, storage, and deployment behavior using supplied evidence.
- Prepare a Root Cause Review: Turn incident evidence into causal hypotheses, corrective actions, monitoring improvements, and postmortem inputs.

## Prompt Templates

### Assess a Service Problem

```
Help me diagnose [symptom] in [environment]. Ask for missing context, then provide a prioritized checklist with safe verification steps.
```

### Investigate Kubernetes Restarts

```
Analyze these pod events, logs, and resource metrics for [workload]. Rank likely causes and propose low-impact tests for each hypothesis.
```

### Correlate Distributed Signals

```
Correlate these logs, metrics, and traces across [services]. Identify the failure path, evidence gaps, and the safest remediation sequence.
```

### Lead Incident Resolution

```
Create an incident plan for [impact and scope]. Separate containment, recovery, root cause validation, stakeholder updates, and long-term prevention.
```

## Limitations

- It cannot inspect live systems unless the user provides data or grants compatible tool access.
- Recommendations require environment-specific validation before production use.
- It does not replace organizational incident procedures, security controls, or expert approval.
- The referenced implementation playbook is not included, so detailed examples may be unavailable.

## Best Practices

- Provide timestamps, recent changes, impact, architecture context, and representative evidence.
- Validate one hypothesis at a time and record expected versus observed results.
- Review every command and remediation step for permissions, scope, and production impact.

## Anti Patterns

- Do not share secrets, private keys, access tokens, or unredacted customer data.
- Do not apply broad configuration changes before confirming the likely cause and rollback path.
- Do not treat plausible hypotheses as confirmed root causes without supporting evidence.

## Security Audit

- Audited at: 2026-07-23T22:14:36.287\+00:00
- Summary: All six static findings are false positives. Markdown path formatting, descriptive PowerShell references, and general troubleshooting language do not execute commands or direct system reconnaissance. No prompt injection, exfiltration intent, or unsafe automation was found in SKILL.md.

## Stats

- Views: 130
- Downloads: 14
- Favorites: 0
- Popularity score: 0