[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M75828B1FMP7W48KMD7X5ZWB":3,"skill-same:01M75828B1FMP7W48KMD7X5ZWB":55},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":29,"repositoryHtmlUrl":30,"repositoryHomepageUrl":31,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M75828B1FMP7W48KMD7X5ZWB","cron-cost-scheduling",null,"Schedule LLM crons around provider peak\u002Foff-peak windows.","lukemcqueen\u002Fsteadfaste-fleet","lukemcqueen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F405240?v=4","01M757MHPAGRXTJAT46SKZP98R","Luke McQueen","01M757MHPZ573KP8DT9GB8SBQ7",4,"MIT","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fcron-cost-scheduling\u002FSKILL.md",[],[],"2026-11-12 13:55:43",false,"---\nname: cron-cost-scheduling\nversion: 1.0.0\ncategory: devops\ndescription: \"Schedule LLM crons around provider peak\u002Foff-peak windows.\"\n---\n\n# Cron Cost Scheduling\n\nWhen a provider introduces time-of-use pricing (peak vs off-peak rates) and the\nuser asks whether LLM crons should be rescheduled to save money. Covers the\nverification, classification, and decision procedure. Does NOT cover cost\nmetering (no cost-capture store ships with the fleet) — this is about\nscheduling decisions.\n\n## When to Use\n\n- User asks \"when are the off-peak hours for \u003Cregion>?\" or \"should we move our LLM crons off-peak?\"\n- A provider announces a pricing change with time-of-use windows\n- Reviewing whether existing cron schedules are cost-optimal\n\n## Procedure\n\n### 1. Fetch the LIVE pricing page — never trust search snippets\n\nSearch results carry stale windows for months. The old DeepSeek \"off-peak\n16:30–00:30 UTC\" card appeared in search results alongside the live card\n(2026-08-18). Always pull the provider's own pricing page and grep the window\ntext:\n\n```bash\ncurl -sL \u003Cpricing-url> -o \u002Ftmp\u002Fprovider_pricing.html\ngrep -oiE '.{80}(peak hours|off-peak|UTC|discount).{80}' \u002Ftmp\u002Fprovider_pricing.html\n```\n\nDocusaurus\u002FNext.js doc pages render fine as static HTML — no browser needed.\nIf the pricing table only appears without its footnote, grep for `utc` and\n`peak` separately; the footnote is usually below the table.\n\n### 2. Convert the window to the user's local time\n\nWindows are always stated in UTC. Seoul = UTC+9. DeepSeek live card\n(2026-08-18): \"Peak hours are 01:00 - 04:00 and 06:00 - 10:00 UTC (all other\nhours are off-peak)\" → Seoul peak = 10:00–13:00 and 15:00–19:00 KST, off-peak\n= 19:00–10:00 plus the 13:00–15:00 lunch gap. State the local windows\nexplicitly in the delivery — don't make the user do the conversion.\n\n### 3. Classify every LLM cron against the windows\n\nExpand each cron's hour\u002Fminute fields (ranges `9-17`, lists `18,20,22`,\nwildcards) and count peak hits per job. Most night\u002Fearly-morning crons already\nrun off-peak — zero change needed; say so instead of proposing mass\nrescheduling. Classification table for the DeepSeek case:\n`references\u002Fdeepseek-peak-offpeak-2026-08.md`.\n\n### 4. Decision rule — move only the free wins\n\n- **Do NOT move crons whose purpose is daytime responsiveness** (workday\n  auto-remediation: `agent-fixer-workday`, `fleet-bus-workday`,\n  `orch-backlog-driver`). Their function beats the pennies (~$0.003–0.007\n  extra per peak run at flash rates). Rescheduling them to off-peak-only\n  guts the service to save cents.\n- **DO move runs sitting just inside the peak boundary** where sliding a few\n  minutes costs zero coverage (e.g. 18:50→19:50, 12:00→13:30). These are the\n  only free wins.\n- Night crons are already off-peak — leave them.\n\n### 5. Quantify before trimming frequency\n\nMeasure before trimming frequency: gather a week of real spend rather than\ntrimming on estimates (no cost-capture store ships with the fleet — use the\nprovider's own usage\u002Fbilling figures when a decision needs numbers). Cron-run spend is often trivial next\nto interactive-session spend — the recommendation should say so.\n\n## Pitfalls\n\n- **Stale search snippets outnumber live pages.** A third-party \"pricing\n  clock\" or aggregator blog can keep the old window for months after a rate\n  card change. The provider's own page is the only authority.\n- **Rate cards change without fanfare.** The DeepSeek peak\u002Foff-peak card was\n  live by 2026-08-16 while older docs pages still implied the old flat\n  pricing. Re-verify the page every time this question comes up.\n- **Don't propose moving daytime crons to night.** It reads as cost rigor but\n  is actually a service regression — call it out explicitly with the\n  per-run cost delta so the user can judge.\n- **web_extract may be search-only on some backends** (DDG); if it errors,\n  fall back to `curl` + grep as in step 1 — the page is static HTML.\n\n## Verification\n\n- The window quote comes from the provider's own page fetched this session\n  (cite the URL and date).\n- The local-time conversion is stated explicitly (e.g. \"UTC+9 → KST\").\n- Every LLM cron's peak-hit count is backed by an actual parse of the live\n  `cronjob action='list'` output.\n- The recommendation names which crons to move, which to keep, and why.\n\n## Related\n\n- `cron-job-management` — naming, install, doctor-sync for any schedule change you do make\n","# Cron Cost Scheduling\n\nWhen a provider introduces time-of-use pricing (peak vs off-peak rates) and the\nuser asks whether LLM crons should be rescheduled to save money. Covers the\nverification, classification, and decision procedure. Does NOT cover cost\nmetering (no cost-capture store ships with the fleet) — this is about\nscheduling decisions.\n\n## When to Use\n\n- User asks \"when are the off-peak hours for \u003Cregion>?\" or \"should we move our LLM crons off-peak?\"\n- A provider announces a pricing change with time-of-use windows\n- Reviewing whether existing cron schedules are cost-optimal\n\n## Procedure\n\n### 1. Fetch the LIVE pricing page — never trust search snippets\n\nSearch results carry stale windows for months. The old DeepSeek \"off-peak\n16:30–00:30 UTC\" card appeared in search results alongside the live card\n(2026-08-18). Always pull the provider's own pricing page and grep the window\ntext:\n\n```bash\ncurl -sL \u003Cpricing-url> -o \u002Ftmp\u002Fprovider_pricing.html\ngrep -oiE '.{80}(peak hours|off-peak|UTC|discount).{80}' \u002Ftmp\u002Fprovider_pricing.html\n```\n\nDocusaurus\u002FNext.js doc pages render fine as static HTML — no browser needed.\nIf the pricing table only appears without its footnote, grep for `utc` and\n`peak` separately; the footnote is usually below the table.\n\n### 2. Convert the window to the user's local time\n\nWindows are always stated in UTC. Seoul = UTC+9. DeepSeek live card\n(2026-08-18): \"Peak hours are 01:00 - 04:00 and 06:00 - 10:00 UTC (all other\nhours are off-peak)\" → Seoul peak = 10:00–13:00 and 15:00–19:00 KST, off-peak\n= 19:00–10:00 plus the 13:00–15:00 lunch gap. State the local windows\nexplicitly in the delivery — don't make the user do the conversion.\n\n### 3. Classify every LLM cron against the windows\n\nExpand each cron's hour\u002Fminute fields (ranges `9-17`, lists `18,20,22`,\nwildcards) and count peak hits per job. Most night\u002Fearly-morning crons already\nrun off-peak — zero change needed; say so instead of proposing mass\nrescheduling. Classification table for the DeepSeek case:\n`references\u002Fdeepseek-peak-offpeak-2026-08.md`.\n\n### 4. Decision rule — move only the free wins\n\n- **Do NOT move crons whose purpose is daytime responsiveness** (workday\n  auto-remediation: `agent-fixer-workday`, `fleet-bus-workday`,\n  `orch-backlog-driver`). Their function beats the pennies (~$0.003–0.007\n  extra per peak run at flash rates). Rescheduling them to off-peak-only\n  guts the service to save cents.\n- **DO move runs sitting just inside the peak boundary** where sliding a few\n  minutes costs zero coverage (e.g. 18:50→19:50, 12:00→13:30). These are the\n  only free wins.\n- Night crons are already off-peak — leave them.\n\n### 5. Quantify before trimming frequency\n\nMeasure before trimming frequency: gather a week of real spend rather than\ntrimming on estimates (no cost-capture store ships with the fleet — use the\nprovider's own usage\u002Fbilling figures when a decision needs numbers). Cron-run spend is often trivial next\nto interactive-session spend — the recommendation should say so.\n\n## Pitfalls\n\n- **Stale search snippets outnumber live pages.** A third-party \"pricing\n  clock\" or aggregator blog can keep the old window for months after a rate\n  card change. The provider's own page is the only authority.\n- **Rate cards change without fanfare.** The DeepSeek peak\u002Foff-peak card was\n  live by 2026-08-16 while older docs pages still implied the old flat\n  pricing. Re-verify the page every time this question comes up.\n- **Don't propose moving daytime crons to night.** It reads as cost rigor but\n  is actually a service regression — call it out explicitly with the\n  per-run cost delta so the user can judge.\n- **web_extract may be search-only on some backends** (DDG); if it errors,\n  fall back to `curl` + grep as in step 1 — the page is static HTML.\n\n## Verification\n\n- The window quote comes from the provider's own page fetched this session\n  (cite the URL and date).\n- The local-time conversion is stated explicitly (e.g. \"UTC+9 → KST\").\n- Every LLM cron's peak-hit count is backed by an actual parse of the live\n  `cronjob action='list'` output.\n- The recommendation names which crons to move, which to keep, and why.\n\n## Related\n\n- `cron-job-management` — naming, install, doctor-sync for any schedule change you do make\n",{"name":5,"version":24,"category":25,"description":7},"1.0.0","devops","main","162463c5db5a1e06f4c54abddce6ee13c3d7612b","skills\u002Fdevops\u002Fcron-cost-scheduling",1,"https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet","steadfaste.com","https:\u002F\u002Fgithub.com\u002Flukemcqueen","AI Architect | Building Hermes Cortex: governed multi-agent AI for autonomous software engineering | Enterprise Tech Consultant","2026-11-14 22:33:34",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add lukemcqueen\u002Fsteadfaste-fleet --skill cron-cost-scheduling",{"tool":41,"command":42},"bunx","bunx skills add lukemcqueen\u002Fsteadfaste-fleet --skill cron-cost-scheduling",{"tool":44,"command":45},"pnpm","pnpm dlx skills add lukemcqueen\u002Fsteadfaste-fleet --skill cron-cost-scheduling","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fcron-cost-scheduling\u002FSKILL.md",true,[49],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"references\u002Fdeepseek-peak-offpeak-2026-08.md","REFERENCE","参考文档",2606,"https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fcron-cost-scheduling\u002Freferences\u002Fdeepseek-peak-offpeak-2026-08.md",[56,64,71,78,85,92],{"skillNo":57,"name":58,"nameZh":6,"description":59,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":60,"categories":61,"keywords":62,"publishedAt":63},"01M7DXY4G8NSVB59K8B6S2AQNP","macbook-linux-maintenance","Legacy MacBook Linux fixes: GPU wedge, reboot hang, thermal.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fmacbook-linux-maintenance\u002FSKILL.md",[],[],"2026-11-15 14:19:17",{"skillNo":65,"name":66,"nameZh":6,"description":67,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":68,"categories":69,"keywords":70,"publishedAt":63},"01M7DXXYJCXFJ3TGB84ZQ8KFE6","governance-regression-triage","Triage a host's governance issues: what changed.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fgovernance-regression-triage\u002FSKILL.md",[],[],{"skillNo":72,"name":73,"nameZh":6,"description":74,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":75,"categories":76,"keywords":77,"publishedAt":63},"01M7DXXSAP1G8FDRJAMXQGBBMK","enforcement-immutability","Use when enforcement files lose their immutable flag.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fenforcement-immutability\u002FSKILL.md",[],[],{"skillNo":79,"name":80,"nameZh":6,"description":81,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":82,"categories":83,"keywords":84,"publishedAt":63},"01M78SSDQVRYX64Y16X32TC0NS","steadfaste-fleet-repo-deploy","Use when committing to the steadfaste-fleet repo.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fsteadfaste-fleet-repo-deploy\u002FSKILL.md",[],[],{"skillNo":86,"name":87,"nameZh":6,"description":88,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":89,"categories":90,"keywords":91,"publishedAt":63},"01M78SS928F3R2RJAYVW4E30CE","steadfaste-fleet-maintenance","Maintain an installed Steadfaste Fleet instance — update both the upstream Hermes Agent and the fleet repo layer, merge diverged histories, sync skills safely, and troubleshoot common issues. Covers the dual-update workflow (hermes update + fleet-update.sh), encrypted brain data backup, mycortex import, and recovery techniques.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Fsteadfaste-fleet-maintenance\u002FSKILL.md",[],[],{"skillNo":93,"name":94,"nameZh":6,"description":95,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":63},"01M78SP1CHW0T0BWE5ZT6JBPX8","fleet-preflight","DEPRECATED alias — merged into survey-before-action (2026-08-20). Load survey-before-action instead; its \"Repo-Specific Pre-Flight\" section covers git search, Hermes boundary, and deployment verification.","https:\u002F\u002Fgithub.com\u002Flukemcqueen\u002Fsteadfaste-fleet\u002Fblob\u002Fmain\u002Fskills\u002Fdevops\u002Ffleet-preflight\u002FSKILL.md",[],[]]