cron-diagnostics

Diagnose cron job health, scheduling conflicts, and cost, then output optimization actions.

6|1|Updated May 11, 2026
One-click install
npx skills add https://github.com/yakeworld/Synthos --skill cron-diagnostics-yakeworld
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: cron-diagnostics
Source: https://github.com/yakeworld/Synthos/tree/main/skills/private/devops/cron-diagnostics
Command: npx skills add https://github.com/yakeworld/Synthos --skill cron-diagnostics-yakeworld

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Cron job lists accumulate paused, erroring, overlapping, and overpriced tasks over time, silently wasting quota and losing results. This Skill audits the full cron task inventory and produces a structured diagnostic report with concrete, executable optimization actions. ## Core Features & Use Cases - Health Diagnosis Matrix: Flags paused tasks, error states, misconfigured no_agent/model combinations, deliver=local result loss, and over-frequent schedules with severity levels. - Overlap & Frequency Clustering: Groups tasks by functional keywords (paper pipeline, evolution, monitoring) and execution frequency to detect mergeable duplicates. - Cost Control: Tracks paid (DeepSeek) tasks and recommends downgrades or merges to keep paid tasks within a 4-6 budget. - Use Case: Run it against a 15-task cron list to get a JSON report plus 5 actionable suggestions (remove/update/merge), reducing the list to 10 tasks and paid tasks from 6 to 4. ## Quick Start Ask the agent to run a cron health diagnostic on all scheduled tasks and produce an optimization plan with executable update, remove, or create actions.

Frequently Asked Questions about cron-diagnostics

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I audit cron jobs for errors and wasted spend?

List all cron tasks, then evaluate each for paused state, last_status errors, deliver configuration, schedule frequency, and paid provider usage. The diagnostic matrix assigns severity levels and produces optimization actions such as remove, update, or merge.

How to detect overlapping or duplicate cron tasks?

Group tasks by functional keywords in their names and descriptions, such as paper pipeline, evolution, or monitoring clusters. Tasks in the same group using similar models and doing similar work are flagged as overlap and recommended for merging.

Why do no_agent cron tasks time out in batches?

Batch timeouts across multiple no_agent script tasks usually indicate a missing codex profile rather than task logic errors. Check the profiles directory with ls ~/.codex/profiles/, restore the missing config, verify connectivity, then resume batch operations.

What cron schedule frequency is reasonable for monitoring tasks?

Heartbeat tasks can run every 30 minutes, scanning tasks should run at most every 4 hours, and paper-processing tasks only need daily schedules. Non-heartbeat tasks running more often than hourly are flagged for frequency reduction.

How many paid model cron tasks should I keep?

Keep paid tasks, such as those using DeepSeek, within a 4-6 task budget and run everything else on free models. Downgrade or merge low-value paid tasks when the count exceeds this range.