Clopus-watcher

lastmod 2026-01-31created 2026-01-311 edits

An autonomous 24/7 on-call engineer in the style of Claude Code - an AI-based monitoring tool aimed at automating incident response in Kubernetes environments

About

• An AI-based monitoring tool aimed at automating incident response in Kubernetes environments
• Uses Claude Code to analyze Pod status in real time, detect errors, and either fix them automatically or generate reports
• Runs periodically as a CronJob, maintaining a 24-hour autonomous response setup even when no operator is around
• Main operations
  • Monitors Pod status in the target Namespace
  • Detects abnormal states such as CrashLoopBackOff and Error
  • Identifies the cause of errors through log analysis
  • Enters the Pod and applies hotfixes when needed
  • Records fix history in an SQLite DB and provides reports
• Supports visualizing error and fix history in a web UI through a separate Dashboard deployment
• Lightweight structure built with Go + HTML + Shell, optimized for Kubernetes CronJob deployment
• MIT license

See also

• Claude Code
• Kubernetes

Favorite site

• Github - kubeden/clopus-watcher
• clopus-watcher - An autonomous 24/7 on-call engineer in the style of Claude Code | GeekNews