Manage SUE-tagged cron entries across Unix, Codex, Kimi, and RunPod backends.
npx skills add https://github.com/dongzhuoyao/deepresearch --skill sue-cronjob
Please help me install this Agent Skill. Skill: sue-cronjob Source: https://github.com/dongzhuoyao/deepresearch/tree/main/.codex/skills/sue-cronjob Command: npx skills add https://github.com/dongzhuoyao/deepresearch --skill sue-cronjob
Managing recurring scheduled tasks for SUE scale-up ML/HPC experiments is error-prone when juggling multiple experiment runs across different backends, as manual cron edits can accidentally disrupt unrelated system jobs or miss critical monitor and cleanup triggers.
Use the sue-cronjob skill to list all active SUE-tagged cron jobs for your current experiment workspace.
High-intent search queries and answers about installing and using this skill.
To safely manage cron jobs for ML experiments, you can use tagged entry validation to only modify entries explicitly tagged with SUE_CRONJOB id=<id>, preventing accidental changes to unrelated system scheduler tasks.
Yes, you can schedule RunPod cleanup and monitor tasks across multiple backends, handling local Unix cron, Codex CLI, Kimi CronCreate, RunPod, and macOS LaunchAgents from a single workflow interface.
You ensure consistent timezone and secret redaction for Codex CLI cron wrappers by enforcing guardrails that apply a default Asia/Shanghai timezone and automatically redact secrets during entry creation.
The best way to track cron job execution status for HPC scale-up experiments is by writing mandatory status updates to a Turso cronjob_runs table, providing direct dashboard observability for your runs.
No, disabling a monitor cron entry does not permanently delete it from Unix cron; the system enforces a disable-instead-of-delete guardrail for temporary stops, allowing you to re-enable it later.
SUE experiment cron jobs require writing to a Turso cronjob_runs table to enforce mandatory dashboard observability, ensuring that every scheduled monitor, dryrun, fullrun, and cleanup execution is tracked.
demithrasManage skill lifecycles with CREATE, IMPROVE, and ASSESS workflows.
niznik-devOrchestrate SLURM fine-tuning jobs with torchtune and evaluate with inspect-ai.
caiokfManage parallel background commands with gob job IDs and output capture.
az9713Convert natural language scheduling requests into cron expressions and crontab entries.
