Cron Health Check

Monitors OpenClaw cron job health, identifies failures, timeouts, and delivery issues.

Security Scan

The skill's code and instructions are coherent with its stated purpose (monitoring OpenClaw cron job health); it reads local OpenClaw job/run files, makes no network calls, and requests no credentials.

Purpose & Capability

The name/description match the implementation: the bundled Python script reads OpenClaw cron configuration (jobs.json) and per-job run logs to identify consecutive failures, timeouts, and delivery issues. The SKILL.md's claims (reports, suggestions like --best-effort-deliver, detection of OpenRouter API limit messages) are implemented via string checks in the script and are proportionate to the purpose. Minor note: SKILL.md shows an example payload referencing an LLM model (openrouter/google/gemini-2.5-flash) — the script itself does not call any model or require model credentials.

Instruction Scope

Runtime instructions tell the user to execute the included script; the example command hardcodes a user path (/Users/ghost/.openclaw/...), which is a benign but user-specific example and should be adjusted. The script reads only OpenClaw-local paths (~/.openclaw/cron/jobs.json and runs/*.jsonl) and prints or JSON-outputs a summary. It does not reach out to external endpoints or attempt to read unrelated system files or environment variables. If you schedule the sample agentTurn payload from SKILL.md, an agent invocation could trigger other services (e.g., an LLM) — that scheduling example is separate from what the script itself does.

Install Mechanism

No install spec and no remote downloads; the skill is instruction-only with a bundled Python script. Nothing is written to disk by an installer and there are no third-party packages fetched by the skill at install time.

Credentials

The skill declares no required environment variables, credentials, or config paths. The Python script uses a default OpenClaw home directory (Path.home() / '.openclaw') or an explicit --openclaw-home argument; it does not read secrets or other env vars. Note: SKILL.md's example scheduling payload references a model provider (openrouter) — if you use that scheduling approach, the runtime that executes the agentTurn might require credentials unrelated to this script.

Persistence & Privilege

The skill does not request persistent/always-on privileges (always is false) and does not modify other skills or global agent configurations. It is user-invocable and can be scheduled, which is appropriate for a monitoring tool.

Assessment

This skill appears to do exactly what it says: it reads OpenClaw's local cron job config and run logs and emits a health summary. Before installing or scheduling it: 1) Update the example invocation path in SKILL.md (/Users/ghost) to your actual location or run the script with --openclaw-home to avoid running the wrong file. 2) Verify that jobs.json and the runs/*.jsonl files it reads do not contain sensitive data you don't want exposed to the agent/user who will run the script. 3) If you plan to use the SKILL.md agentTurn cron payload, be aware that executing that payload may invoke an LLM (the example references an OpenRouter model) — such invocations require separate credentials and may incur cost or external network activity; the included script itself does not perform network calls. 4) Run the script manually first to confirm output and that the default path points to your OpenClaw installation. Overall this skill is internally consistent and low-risk, but confirm the file paths and scheduling behavior in your environment before enabling automated runs.

Cron Health Check | OpenClaw Skill

Description

Monitors OpenClaw cron job health, identifies failures, timeouts, and delivery issues.

Cron Health Check | OpenClaw Skill

Monitors the health of OpenClaw cron jobs by analyzing run history and identifying patterns of failures, timeouts, and delivery issues.

Usage

• As a scheduled cron job to monitor job health proactively
• Manually to check current cron job status
• After fixing cron job errors to verify improvements

# Check health of all cron jobs (last 24 hours)
python3 /Users/ghost/.openclaw/workspace/skills/cron-health-check/scripts/check_cron_health.py

# Check last 48 hours
python3 /Users/ghost/.openclaw/workspace/skills/cron-health-check/scripts/check_cron_health.py --hours 48

# Output JSON format
python3 /Users/ghost/.openclaw/workspace/skills/cron-health-check/scripts/check_cron_health.py --json

What this skill does

• Analyzes cron job run history from the last N hours
• Identifies jobs with consecutive failures, timeouts, or delivery issues
• Reports health status (healthy/warning/critical) for each job
• Suggests fixes (e.g., adding --best-effort-deliver for delivery failures)
• Detects OpenRouter API limit issues

Integration as a Cron Job

This skill can run periodically to monitor cron job health:

{
  "payload": {
    "kind": "agentTurn",
    "message": "Run cron-health-check skill to analyze cron job health.",
    "model": "openrouter/google/gemini-2.5-flash",
    "thinking": "low",
    "timeoutSeconds": 60
  },
  "schedule": {
    "kind": "cron",
    "cron": "0 */6 * * *"
  },
  "delivery": {
    "mode": "announce",
    "bestEffort": true
  },
  "sessionTarget": "isolated",
  "name": "Cron Health Monitor"
}

Output

Health status for each job:

• healthy: No issues detected
• warning: Some issues but not critical
• critical: Multiple consecutive failures or timeouts

Issues detected:

• Consecutive errors
• Timeout patterns
• Delivery failures (suggests --best-effort-deliver)
• OpenRouter API limit exceeded

Exit Codes

• 0: All jobs healthy
• 1: Warning issues found
• 2: Critical issues found