openclaw-guardian and openclaw-agent-doctor
These are complements—Guardian watchdog serves as the monitoring and orchestration layer that automatically invokes the Agent Doctor's diagnostic and repair functions when issues are detected in the OpenClaw Gateway infrastructure.
About openclaw-guardian
LeoYeAI/openclaw-guardian
🛡️ Guardian watchdog for OpenClaw Gateway — auto-monitor, self-repair via doctor --fix, git-based rollback, daily snapshots, and Discord alerts. Powered by MyClaw.ai
Implements a cascading recovery strategy with exponential escalation: failed health checks trigger automated repair attempts via `doctor --fix`, then escalate to git-based workspace rollback if repairs exhaust, with configurable cooldown periods to prevent thrashing. Integrates with OpenClaw Gateway's diagnostic tooling and operates alongside existing watchdog processes, offering complementary coverage through daily git snapshots and optional Discord webhook notifications for alert aggregation.
About openclaw-agent-doctor
AlekseiUL/openclaw-agent-doctor
🏥 Self-diagnostic toolkit for OpenClaw AI agents. Find and fix problems before they find you.
Implements comprehensive system health checks across 7 categories (memory, crons, config, files, gateway, system, security) with a curated database of 28+ common problems and their concrete solutions. Operates as an OpenClaw skill that performs read-only diagnostics with user-confirmed fixes, supporting risk assessment for each remediation, and includes a standalone bash script option for CI/CD integration without requiring agent interaction.
Related comparisons
Scores updated daily from GitHub, PyPI, and npm data. How scores work