AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT-0 Self-run

Incident Triage

skill-aws-samples-sample-aws-ops-skills-for-agents-incident-triage · by aws-samples

>

No reviews yet
0 installs
22 views
0.0% view→install

Install

$ agentstack add skill-aws-samples-sample-aws-ops-skills-for-agents-incident-triage

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-aws-samples-sample-aws-ops-skills-for-agents-incident-triage)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Incident Triage? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Incident Triage — 快速分诊

前 5 分钟的决策框架:分类 → 定界 → 路由。不做深度诊断。

和其他 skills 的关系

  • incident-triage 是入口——快速判断后路由到具体 skill
  • SEV1/2 诊断 → aws-troubleshooting(深度排查)
  • 需要看日志 → log-diagnosis
  • 需要全面巡检 → health-check

工作流程

Step 1 — 收集信号(< 1 分钟)

MUST:

  • 如果是告警触发:
  • 读告警详情:aws cloudwatch describe-alarms --alarm-names {name} --region {region}
  • 记录 metric, threshold, duration, current value
  • 如果是用户报告:
  • 确认症状(什么不工作了)
  • 确认开始时间(什么时候开始的)
  • 确认影响范围(谁受影响)
  • 快速查当前告警全景:
  • aws cloudwatch describe-alarms --state-value ALARM --region {region}
  • 对账号的主要 region 都查一遍
  • 查最近 15 分钟的变更事件:
  • aws cloudtrail lookup-events --lookup-attributes AttributeKey=ReadOnly,AttributeValue=false --start-time {T-15m} --region {region}

SHOULD:

  • 检查 AWS Health Dashboard:aws health describe-events --filter '{"eventStatusCodes":["open","upcoming"]}' --region us-east-1
  • 如果有 dashboard,快速看关键指标趋势

Step 2 — 判定 Severity

MUST:

  • references/severity-matrix.md 做分类
  • 基于证据判断,不基于猜测

分类速查:

  • SEV1:服务完全不可用 / 数据风险 → 立即行动
  • SEV2:服务降级 / 部分受影响 → 15 分钟内行动
  • SEV3:异常观察 / 不影响服务 → 1 小时内行动

Step 3 — 定界影响范围

MUST:

  • 判断 blast radius:单资源 → 单服务 → 单 AZ → 单 region → 跨 region
  • 检查关联服务(参考 aws-troubleshooting/references/investigation-framework.md 的依赖链表)

SHOULD:

  • 用 health-check 做快速全局巡检确认是否有连锁影响
  • 检查是否是 AWS 侧事件(Health Dashboard)

Step 4 — 路由 + 输出

MUST:

  • 输出 incident report(assets/incident-report-template.md 格式)
  • 包含:severity, blast radius, 信号摘要, 初步判断, 建议操作

路由规则:

  • SEV1 → 路由到 aws-troubleshooting(实时查 aws-knowledge MCP + SSM Automation runbooks)
  • SEV2 + 需要日志 → 路由到 log-diagnosis
  • SEV2 + 需要指标 → 路由到 health-check
  • SEV3 → 路由到 health-check 做全面巡检

SHOULD:

  • 对 SEV1/2 提供具体的缓解命令(但标记 human-required,不自动执行)

查询参考

  • Severity 分类规则:references/severity-matrix.md
  • 跨服务依赖链:../aws-troubleshooting/references/investigation-framework.md
  • 具体服务排查 SOP:aws-troubleshooting skill(实时查 aws-knowledge MCP)

Automation Boundary

auto-safe

  • aws cloudwatch describe-alarms(查告警状态)
  • aws cloudtrail lookup-events(查最近变更)
  • aws health describe-events(查 AWS 侧事件)
  • describe / list / get 类资源查询
  • 生成 incident report

human-required

  • 所有缓解/修复操作(重启 pod、扩容、回滚、修改配置)
  • 开 AWS Support Case
  • 通知其他人

Gotchas

  • CloudTrail 事件有 ~5 分钟延迟,最近的变更可能还没出现
  • aws health API 只在 us-east-1 有全局事件,其他 region 只有 region-specific 事件
  • INSUFFICIENT_DATA 告警不一定是问题——可能是新创建的告警还没收到足够数据
  • 部署期间(rolling update)的短暂告警是正常的,要看是否在部署窗口内
  • 不要在 SEV1 上花超过 5 分钟做分析——如果 5 分钟内找不到根因,升级

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.