[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79EN6KX9R84WKAJZMS7MHV9":3,"skill-same:01M79EN6KX9R84WKAJZMS7MHV9":62},{"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":24,"commitSha":25,"skillDirectory":26,"forksCount":14,"repositoryHtmlUrl":27,"repositoryHomepageUrl":6,"ownerHtmlUrl":28,"ownerBlogUrl":29,"ownerDescription":30,"updatedAt":31,"install":32,"resources":45},"01M79EN6KX9R84WKAJZMS7MHV9","development-steward",null,"Run the scheduled, read-only weekly maintenance review for the six explicitly opted-in repositories and prepare one concise private report for Check-in.","AO-HyS\u002Fdevelopment-system","AO-HyS","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F286554351?v=4","01M79DCXZGC6076EH5QATNXCY2","AO HyS","01M79DCY06TSBCJ946VP0BT0RG",0,"MIT","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: development-steward\ndescription: Run the scheduled, read-only weekly maintenance review for the six explicitly opted-in repositories and prepare one concise private report for Check-in.\n---\n\n# Development Steward\n\nRun headlessly every week. The initial allowlist is exactly AO HyS, Casa Roca, The Barber Central, NutriPlan, ETERIA, and Development System. Never add another repository without explicit opt-in.\n\nT3 scheduler is the preferred current-host runner. Audit both T3 scheduled tasks and the existing macOS LaunchAgent before activation: retain exactly one recurring runner, never activate a duplicate. Keep one stable task and return results to its owning thread. Monday 09:00 local is the existing cadence; activation requires explicit authorization. Legacy launchd activation is retired even from an ordinary terminal. Only `development-steward-schedule-audit` and `development-steward-schedule-disable` remain for inspection and disabling; preserve historical reports and the durable retro ledger. No schedule is enabled by loading this skill.\n\nRequest Sol 6.1 High for collection and computer use; report observed identity or unknown, never infer it from the request. Quota recovery uses existing bounded authorized profile recovery and receipts, with no silent provider change.\n\nFor Development System, Casa Roca, Barber and NutriPlan, inspect up to ten recent relevant real sessions per repository. Supply stable `repositoryId\u002FsessionId`, exact revision or updatedAt, `complete`, and bounded evidence pointers; never raw transcripts or credentials. Use `development-system development-steward --input \u003Cprivate-json> --home \u003Chome> --json` to reconcile the private atomic ledger. Complete session collection advances only its cursor; recommendations remain pending until a verified PR receipt or explicit resolved\u002Fno-change. Replaying a processed session reports already-processed. PR identity is repository plus stable changeId across revisions; review an existing open PR instead of creating a duplicate. Failed publication stays pending and retryable. Unknown statuses stay unproven.\n\nDraft preparation requires proven revision, explicit clean=true and activeWriter=false, deterministic safeUpdate and focused checks. Missing ownership or cleanliness evidence blocks drafts. Limit to one bounded draft per repository. The legacy `evaluations` field contains observations only and never authorizes evals or automated tests.\n\nInspect skills, Codex Security, React, TanStack, shadcn, Convex, Cloudflare, Expo\u002Fmobile, PostHog, Release Train, and repository-declared fitness functions. Always treat React Doctor, the Impeccable CLI, the Impeccable umbrella skill, and the Matt Pocock skill bundle as four explicit upstream checks rather than hiding them inside a generic `skills` row. Compare pinned upstream versions using both an exact diff and changelog. Never adopt `latest`, and never use stars as the only adoption signal.\n\nFor the Matt Pocock bundle, verify both provenance and behavior: the installed `grilling` skill must ask the whole current question frontier in one numbered round and give a recommendation for every question. A version label alone is insufficient evidence. For Impeccable, report the CLI and skill versions independently because they use separate release lines.\n\nRead repeated mistakes with `development-system mistake list --repeated --json`. The report carries them as `repeatedMistakes: [{ id, incidents, proposedControl }]`. For each id, add one bounded item that proposes the hard control at the highest level of the Gardener ladder (code > lint\u002FCI\u002Fguard > rule\u002Fskill), or states the existing control when one is recorded. Never propose a style note alone.\n\nRun `primitives-gardener` in mode tend for each allowlisted repository that has\n`docs\u002Farchitecture\u002Fprimitives.json`, and read `development-system friction list\n--json`. Each primitive drift or friction becomes one pending recommendation in\nthe ledger, deduplicated and closed only by evidence. Frictions that need a\nproduct decision carry a recommendation for the owner; they never count toward\nthe repeated-mistake rule.\n\nCheck `development-system pr-review-status` for PRs merged since the last run;\na merged head with open Codex findings is an action item.\n\nDiscovery is read-only. Keep repository failures local and report missing or stale evidence as unproven. Produce one short private report with what changed, what remains healthy, what needs action, links\u002Fevidence, and whether each human action is mobile or computer work. Feed that report into Check-in.\n\nA deterministic safe update may prepare a branch and draft PR after focused checks. It must never auto-merge, release, promote production, run destructive migrations, or delete tracker state. Those operations keep their separate human authorization.\n","# Development Steward\n\nRun headlessly every week. The initial allowlist is exactly AO HyS, Casa Roca, The Barber Central, NutriPlan, ETERIA, and Development System. Never add another repository without explicit opt-in.\n\nT3 scheduler is the preferred current-host runner. Audit both T3 scheduled tasks and the existing macOS LaunchAgent before activation: retain exactly one recurring runner, never activate a duplicate. Keep one stable task and return results to its owning thread. Monday 09:00 local is the existing cadence; activation requires explicit authorization. Legacy launchd activation is retired even from an ordinary terminal. Only `development-steward-schedule-audit` and `development-steward-schedule-disable` remain for inspection and disabling; preserve historical reports and the durable retro ledger. No schedule is enabled by loading this skill.\n\nRequest Sol 6.1 High for collection and computer use; report observed identity or unknown, never infer it from the request. Quota recovery uses existing bounded authorized profile recovery and receipts, with no silent provider change.\n\nFor Development System, Casa Roca, Barber and NutriPlan, inspect up to ten recent relevant real sessions per repository. Supply stable `repositoryId\u002FsessionId`, exact revision or updatedAt, `complete`, and bounded evidence pointers; never raw transcripts or credentials. Use `development-system development-steward --input \u003Cprivate-json> --home \u003Chome> --json` to reconcile the private atomic ledger. Complete session collection advances only its cursor; recommendations remain pending until a verified PR receipt or explicit resolved\u002Fno-change. Replaying a processed session reports already-processed. PR identity is repository plus stable changeId across revisions; review an existing open PR instead of creating a duplicate. Failed publication stays pending and retryable. Unknown statuses stay unproven.\n\nDraft preparation requires proven revision, explicit clean=true and activeWriter=false, deterministic safeUpdate and focused checks. Missing ownership or cleanliness evidence blocks drafts. Limit to one bounded draft per repository. The legacy `evaluations` field contains observations only and never authorizes evals or automated tests.\n\nInspect skills, Codex Security, React, TanStack, shadcn, Convex, Cloudflare, Expo\u002Fmobile, PostHog, Release Train, and repository-declared fitness functions. Always treat React Doctor, the Impeccable CLI, the Impeccable umbrella skill, and the Matt Pocock skill bundle as four explicit upstream checks rather than hiding them inside a generic `skills` row. Compare pinned upstream versions using both an exact diff and changelog. Never adopt `latest`, and never use stars as the only adoption signal.\n\nFor the Matt Pocock bundle, verify both provenance and behavior: the installed `grilling` skill must ask the whole current question frontier in one numbered round and give a recommendation for every question. A version label alone is insufficient evidence. For Impeccable, report the CLI and skill versions independently because they use separate release lines.\n\nRead repeated mistakes with `development-system mistake list --repeated --json`. The report carries them as `repeatedMistakes: [{ id, incidents, proposedControl }]`. For each id, add one bounded item that proposes the hard control at the highest level of the Gardener ladder (code > lint\u002FCI\u002Fguard > rule\u002Fskill), or states the existing control when one is recorded. Never propose a style note alone.\n\nRun `primitives-gardener` in mode tend for each allowlisted repository that has\n`docs\u002Farchitecture\u002Fprimitives.json`, and read `development-system friction list\n--json`. Each primitive drift or friction becomes one pending recommendation in\nthe ledger, deduplicated and closed only by evidence. Frictions that need a\nproduct decision carry a recommendation for the owner; they never count toward\nthe repeated-mistake rule.\n\nCheck `development-system pr-review-status` for PRs merged since the last run;\na merged head with open Codex findings is an action item.\n\nDiscovery is read-only. Keep repository failures local and report missing or stale evidence as unproven. Produce one short private report with what changed, what remains healthy, what needs action, links\u002Fevidence, and whether each human action is mobile or computer work. Feed that report into Check-in.\n\nA deterministic safe update may prepare a branch and draft PR after focused checks. It must never auto-merge, release, promote production, run destructive migrations, or delete tracker state. Those operations keep their separate human authorization.\n",{"name":5,"description":7},"main","9f5b43aadb1a92441d65e9cb74a584cd6fca62bc","artifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system","https:\u002F\u002Fgithub.com\u002FAO-HyS","https:\u002F\u002Faohys.net\u002F","Software Boutique, we make your projects come true live. ","2026-11-14 03:17:27",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add AO-HyS\u002Fdevelopment-system --skill development-steward",{"tool":38,"command":39},"bunx","bunx skills add AO-HyS\u002Fdevelopment-system --skill development-steward",{"tool":41,"command":42},"pnpm","pnpm dlx skills add AO-HyS\u002Fdevelopment-system --skill development-steward","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002FSKILL.md",true,[46,52,56],{"path":47,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":50,"sourceHtmlUrl":51},"agents\u002Fopenai.yaml","REFERENCE","参考文档",248,"https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002Fagents\u002Fopenai.yaml",{"path":53,"resourceType":48,"resourceTypeDesc":49,"sizeBytes":54,"sourceHtmlUrl":55},"references\u002Fprompt.md",5572,"https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002Freferences\u002Fprompt.md",{"path":57,"resourceType":58,"resourceTypeDesc":59,"sizeBytes":60,"sourceHtmlUrl":61},"scripts\u002Frunner.mjs","SCRIPT","脚本",6925,"https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.3.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002Fscripts\u002Frunner.mjs",[63,70,77,82,89,96],{"skillNo":64,"name":65,"nameZh":6,"description":66,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M79EQ4BPYQTX8RWN1Y1J2K1F","working-backwards","Define a software feature through PRFAQ and staged documents when the user requests working backwards before coding.","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.8.0\u002Fskills\u002Finternal\u002Fworking-backwards\u002FSKILL.md",[],[],{"skillNo":71,"name":72,"nameZh":6,"description":73,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":74,"categories":75,"keywords":76,"publishedAt":19},"01M79EQ2E8P85Y1Y7VNJZMNEJE","flow-implement","Implement a requested software change through verification, corrections, and its authorized delivery endpoint.","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.8.0\u002Fskills\u002Finternal\u002Fflow-implement\u002FSKILL.md",[],[],{"skillNo":78,"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":79,"categories":80,"keywords":81,"publishedAt":19},"01M79EQ0PNMKHPJ78F8WMNX06P","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.8.0\u002Fskills\u002Finternal\u002Fdevelopment-steward\u002FSKILL.md",[],[],{"skillNo":83,"name":84,"nameZh":6,"description":85,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":86,"categories":87,"keywords":88,"publishedAt":19},"01M79EPXJ56MX7955GWP1HD4B0","grill-with-docs","Run one recoverable interview for functional, visual, or mixed software grills; reuse equivalent answers and offer the shared HTML questionnaire.","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.7.0\u002Fskills\u002Finternal\u002Fgrill-with-docs\u002FSKILL.md",[],[],{"skillNo":90,"name":91,"nameZh":6,"description":92,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":93,"categories":94,"keywords":95,"publishedAt":19},"01M79EPTFPHYEV40ZG52Z0GVS2","global-agent-guardrails","Install, audit, test, or update the Development System's global destructive-command guard for Codex, Claude Code and T3 Code. Use when configuring agent safety hooks or investigating whether dangerous shell operations are blocked.","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.7.0\u002Fskills\u002Finternal\u002Fglobal-agent-guardrails\u002FSKILL.md",[],[],{"skillNo":97,"name":98,"nameZh":6,"description":99,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":100,"licenseFileUrl":101,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":19},"01M79EPR0KPHKAFN7V6CEEDR0Y","impeccable","Design or refine frontend UI using project context and a focused design playbook; also handles visual critique and polish.","Apache-2.0","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.7.0\u002Fskills\u002Fexternal\u002Fimpeccable\u002FLICENSE","https:\u002F\u002Fgithub.com\u002FAO-HyS\u002Fdevelopment-system\u002Fblob\u002Fmain\u002Fartifacts\u002F2.7.0\u002Fskills\u002Fexternal\u002Fimpeccable\u002FSKILL.md",[],[]]