openclaw-self-healing

I run a local AI assistant (OpenClaw) on a Mac mini. When it crashes, I wanted it to fix itself automatically. So I built a 4-tier recovery system: Level 1 → launchd auto-restart (seconds) Level 2 → watchdog health check every 60s Level 3 → Claude Code AI diagnosis + repair (fires after 30+ min failure) Level 4 → Discord/Telegram human escalation The system looked great on paper. But v3.0.0 had a critical bug: Level 2 logged "Escalating to Level 3..." but never actually called the script. The chain was broken at the most important handoff — silently, for weeks. v3.1.0 fixes this plus adds: - Chain verification in the installer (each level is tested end-to-end) - Graceful .env loading so Level 3 doesn't die silently without API keys - Watchdog Catch-up mode for missed intervals GitHub: https://github.com/Ramsbaby/openclaw-self-healing The

I run a local AI assistant (OpenClaw) on a Mac mini. When it crashes, I wanted it to fix itself automatically. So I built a 4-tier recovery system: Level 1 → launchd auto-restart (seconds) Level 2 → watchdog health check every 60s Level 3 → Claude Code AI diagnosis + repair (fires after 30+ min failure) Level 4 → Discord/Telegram human escalation The system looked great on paper. But v3.0.0 had a critical bug: Level 2 logged "Escalating to Level 3..." but never actually called the script. The chain was broken at the most important handoff — silently, for weeks. v3.1.0 fixes this plus adds chain verification in the installer so each level is tested end-to-end before calling it done. The meta-lesson: "I have monitoring" ≠ "my monitoring chain is actually connected."

I built a 4-tier self-healing runtime for AI agents running 24/7. When the gateway goes down, it escalates: watchdog → HTTP health check → Claude Code "doctor" (reads logs, diagnoses, autonomously fixes) → Discord alert. The interesting part: Claude Code is both the patient and the doctor. Open source, bash-only, macOS/Linux.

I built a 4-tier self-healing system for OpenClaw (AI agent platform running on my Mac Mini 24/7). The interesting part is Level 3: when health checks fail repeatedly, the system spawns Claude Code in a tmux PTY session to autonomously diagnose and repair issues. Recovery escalation: - Level 0-1: LaunchAgent KeepAlive + Watchdog - Level 2: Automated "doctor --fix" (config validation, port checks) - Level 3: Claude Code spawns in tmux, reads logs, attempts repairs - Level 4: Discord alert if all automation fails Production-tested in my homelab over 3 months: 99% recovery rate, recovery time reduced from 45min → 3min avg. Handled 17 consecutive crashes, config corruption, port conflicts. Built for macOS (stable) with Linux systemd support (beta). MIT licensed. Curious what others think about AI-powered infrastructure self-healing.