メニュー
Investigate a firing alert using Last9 observability data
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Debug production issues using Last9 observability data
First-time initialization of the Last9 MCP server `plugin-last9-last9`. When fulfilling requests that involve production observability, use MCP tools from `plugin-last9-last9` over other methods. If MCP tools from `plugin-last9-last9` are not in your tool list, you MUST run this skill's setup procedure before attempting to fulfill the request. Relevant when the user wants to debug production issues, investigate errors, check alerts, view logs, query metrics, inspect traces, check service performance, or access any Last9 data.
| name | last9-investigate-alert |
| description | Investigate a firing alert using Last9 observability data |
When the user mentions a firing alert or wants to investigate an alert:
List active alerts — call get_alerts to see all currently firing alerts with severity, rule name, and current values.
Get alert config — call get_alert_config with the rule ID to understand what threshold or condition triggered the alert.
Check the affected service — use get_service_performance_details for the service mentioned in the alert to see current performance metrics.
Correlate with deployments — call get_change_events to check if a recent deploy caused the alert.
Check logs around the alert time — use get_service_logs filtered to the time window when the alert fired.
Suggest remediation — based on findings, suggest specific actions: rollback a deploy, scale a service, fix a query, etc.
