OpenClaw Operator
v1.0.0Diagnose and prevent OpenClaw agent failures — session bloat, lane deadlocks, bootstrap truncation, auth errors, compaction timeouts, and more. Use this skil...
⭐ 0· 112·0 current·0 all-time
byDon Zurbrick@zurbrick
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
The name and description match the delivered artifacts: diagnostic prose, failure-patterns reference docs, and two bash scripts that check OpenClaw runtime files and logs under ~/.openclaw. All required accesses (session files, jobs.json, AGENTS.md, gateway.err.log) are appropriate for an ops/troubleshooting skill.
Instruction Scope
Instructions direct the agent or operator to read and (in break‑glass procedures) edit local OpenClaw state such as ~/.openclaw/cron/jobs.json, session registries, and AGENTS.md. This is expected for an ops tool, but it does grant the skill authority to inspect and advise modification of sensitive local agent state; SKILL.md explicitly allows manual edits as break-glass — users should follow the recommended backup & validation steps before making changes.
Install Mechanism
No install spec is provided (instruction-only with bundled scripts). Nothing is downloaded or installed automatically. The scripts are plain shell/Python and will only be executed if the operator runs them or registers them with OpenClaw cron.
Credentials
No environment variables, credentials, or third‑party tokens are requested. The scripts read local files only. Example CLI snippets reference OpenClaw commands and model names but do not require secret values from the skill itself.
Persistence & Privilege
The skill does not request permanent 'always' inclusion or write to other skills' configuration. It suggests installing cron jobs (normal for ops tooling), but installation is opt‑in. Autonomous model invocation is the platform default and is not changed by this skill.
Assessment
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.Like a lobster shell, security has layers — review code before you run it.
latestvk97a2aks1rme63ctdp5n93gw9s83ddmc
License
MIT-0
Free to use, modify, and redistribute. No attribution required.