本文へスキップ
TakoAPI
スキル一覧に戻る

job-execution-monitor

tradmangh 作

Communication
—まだ評価がありません
あなたの評価

概要

Monitor scheduled jobs (cron) and alert when they fail or miss their schedule.

説明

このスキルでできること

Monitors scheduled cron jobs and alerts when they fail or miss their expected run time. Runs a heartbeat check roughly once per hour, compares each job's last run against its configured schedule and tolerance, and tracks alert state to prevent duplicate notifications.

Reuses existing agent infrastructure and cron tooling instead of requiring a separate monitoring service, keeping daily cost under $0.01 with no additional setup.

使いどころ

  • Catching a nightly database backup that failed to run
  • Detecting a missed daily report generation
  • Confirming overnight batch processes completed on time
  • Alerting when a scheduled data sync stops unexpectedly
  • Verifying a recurring automation ran within its tolerance window

ワークフロー例

入力

User asks: why didn't my daily-wrap-up job run last night?

エージェント
  1. 1

    Load job-execution-monitor.json to get configured jobs and schedules

  2. 2

    Call cron list to retrieve last run times and statuses

  3. 3

    Compare daily-wrap-up last run time against expected schedule plus tolerance

  4. 4

    Detect the job is overdue and send a wake event with expected time and actual last run time

  5. 5

    Record alert in state file to prevent repeated notifications

出力

Alert: 'daily-wrap-up' missed schedule. Expected: 22:00 ±10min. Last run: 5h 32m ago.

インストール

clawhub install tradmangh/job-execution-monitor

CLIコマンドを実行するか、Claude Code、Cursor、Codexなどお使いのコーディングエージェントにプロンプトを貼り付けてください。

リンク