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
✓ PassedNo 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.
Verified badge
Passed review? Show it. Paste this badge into your README, it links to the public security report.
Reliability & compatibility
Declared compatibility
Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.
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 →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:
- 读
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 healthAPI 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.
- Author: aws-samples
- Source: aws-samples/sample-aws-ops-skills-for-agents
- License: MIT-0
Install and usage instructions live in the source repository linked above.
Reviews
No reviews yet, be the first.
Write a review
Versions
- v0.1.0 Imported from the upstream source.