OpenClaw Operator

Diagnose and prevent OpenClaw agent failures — session bloat, lane deadlocks, bootstrap truncation, auth errors, compaction timeouts, and more. Use this skil...

This skill appears to do what it says: read OpenClaw logs/configs and run local diagnostics. Before using it: (1) Review the two included scripts line-by-line to confirm behavior for your environment; they read and parse files under $HOME/.openclaw and call python3. (2) Back up any files the skill suggests editing (e.g., cron/jobs.json, sessions.json, AGENTS.md) before applying changes — the SKILL.md explicitly recommends timestamped backups for break‑glass edits. (3) Note some shell commands assume BSD utilities (stat -f, date -j); test on a non-production host or adapt to Linux stat/date if needed. (4) Run the scripts manually first (bash scripts/session-health-watchdog.sh; bash scripts/bootstrap-budget-check.sh) rather than auto-registering as a cron job until you’ve validated outputs and alerting behavior. (5) If you will register the watchdog as a cron job, ensure the account running cron has appropriate permissions and that you understand where and how alerts will be surfaced (the scripts themselves do not post externally). Overall the skill is internally consistent with its stated ops purpose.

Type: OpenClaw Skill Name: openclaw-operator Version: 1.0.0 The skill bundle is a legitimate administrative toolset for managing and troubleshooting OpenClaw agent installations. It contains diagnostic scripts (session-health-watchdog.sh and bootstrap-budget-check.sh) that monitor session sizes, cron job configurations, and log files for common failure patterns. While the skill requires access to the ~/.openclaw directory and the ability to execute shell commands, its behavior is transparent, well-documented, and strictly limited to the stated purpose of operational maintenance without any signs of data exfiltration, obfuscation, or malicious intent.