[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6M9JJ8SDZ95A5BW6ENSM71C":3,"skill-same:01M6M9JJ8SDZ95A5BW6ENSM71C":69},{"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":40,"linkOnly":41,"markdownContent":42,"markdownBody":43,"frontMatter":44,"refName":45,"commitSha":46,"skillDirectory":47,"forksCount":48,"repositoryHtmlUrl":49,"repositoryHomepageUrl":50,"ownerHtmlUrl":51,"ownerBlogUrl":52,"ownerDescription":53,"updatedAt":54,"install":55,"resources":68},"01M6M9JJ8SDZ95A5BW6ENSM71C","karpathy-guardrails",null,"Behavioral guardrails for Cavekit agents. Four principles — think before coding,\nsimplicity first, surgical changes, goal-driven execution — that prevent\nover-engineering, silent assumptions, scope creep, and unfocused work. Every\ntask-builder, reviewer, planner, and inspector must internalize these before\nwriting a single line. Trigger phrases: \"guardrails\", \"karpathy\", \"scope creep\",\n\"over-engineering\", \"stop adding features\", \"surgical fix\".","hashgraph-online\u002Fawesome-codex-plugins","hashgraph-online","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160371962?v=4","01M6CNB6NRJVNNGN14TR01Q6D3","HOL (Hashgraph Online)","01M6JT1HSKKG9V68KQAQGBA7XP",1554,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FJuliusBrussee\u002Fblueprint\u002Fskills\u002Fkarpathy-guardrails\u002FSKILL.md",[],[19,20,21,22,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39],"Guardrails","karpathy guardrails","karpathy","single","writing","features","for","scope","line","trigger","task","stop","builder","goal","behavioral","simplicity","fix","work","reviewer","coding","execution","2026-11-10 10:12:30",false,"---\nname: karpathy-guardrails\ndescription: |\n  Behavioral guardrails for Cavekit agents. Four principles — think before coding,\n  simplicity first, surgical changes, goal-driven execution — that prevent\n  over-engineering, silent assumptions, scope creep, and unfocused work. Every\n  task-builder, reviewer, planner, and inspector must internalize these before\n  writing a single line. Trigger phrases: \"guardrails\", \"karpathy\", \"scope creep\",\n  \"over-engineering\", \"stop adding features\", \"surgical fix\".\n---\n\n# Karpathy Guardrails\n\nFour rules. Load them into context at the start of every task. The reviewer\nenforces them as a Pass-1 filter before it looks at code quality.\n\n## 1. Think Before Coding\n\nBefore the first edit, write down:\n\n- **What am I actually building?** One sentence. If you cannot state it, stop.\n- **What am I assuming?** List every assumption. If any is load-bearing and\n  unverified, flag `NEEDS_CONTEXT` and ask — do not guess.\n- **What does success look like?** Map each acceptance criterion to a concrete\n  test, check, or observable behaviour. If a criterion is not verifiable,\n  propose a sharpening via the `revision` skill (automated-trace subsection), not a vague attempt.\n\nRefusing to produce code is allowed. A task with unknown scope is a spec bug,\nnot a coding task.\n\n## 2. Simplicity First\n\nThe correct amount of code is the minimum that meets the acceptance criteria.\n\n- No speculative features. No abstraction layer \"in case we need it.\"\n- No new dependencies unless the task requires one and no existing dep fits.\n- No \"while I'm in here\" refactors. Surface them as separate kits.\n- Duplication is not always wrong. Three similar lines usually beat a premature\n  abstraction with two configuration knobs.\n\nIf the diff is larger than the acceptance criteria seem to demand, explain why\nin the commit body. If you cannot, trim the diff.\n\n## 3. Surgical Changes\n\nEvery line in the diff must trace back to an acceptance criterion. Touching\ncode outside the task's owned files is justified only when a requirement forces\nit. Examples of violations:\n\n- Fixing a formatter warning in an unrelated file.\n- Renaming a helper \"to match new convention.\"\n- Reordering imports, docstrings, whitespace.\n- Tightening a type signature the task did not ask about.\n\nIf you see a real bug in adjacent code, log it to `.cavekit\u002Fhistory\u002Fbackprop-log.md`\nas a candidate kit item and keep it out of this task's diff.\n\n## 4. Goal-Driven Execution\n\nTransform vague tasks into verifiable success criteria before execution.\n\n- A task that cannot be verified is not a task — escalate it.\n- The verification plan must be concrete: exact commands, exact assertions,\n  exact files to inspect. \"Make sure it works\" is not a plan.\n- After implementation, run the verification plan. Report the output.\n\n## Role-specific enforcement\n\n- **task-builder** — must produce, alongside code, a Verification Report listing\n  each AC, the verification step, and the observed result.\n- **reviewer** — must refuse to advance to Pass 2 (code quality) if Pass 1 finds\n  any of: undeclared assumptions, diff lines unjustified by an AC, out-of-scope\n  edits, or unreachable verification steps.\n- **planner** — must reject kits that contain un-testable ACs. They are spec\n  bugs and block planning.\n- **inspector** — must flag completed tasks whose verification logs are missing\n  or hand-waved.\n\n## When you are tempted to break a rule\n\nYou are probably over-confident about a shortcut that will cost more than the\ndelay of asking. Stop and note the tension in the commit body or in the build\nlog so the reviewer can judge.\n","# Karpathy Guardrails\n\nFour rules. Load them into context at the start of every task. The reviewer\nenforces them as a Pass-1 filter before it looks at code quality.\n\n## 1. Think Before Coding\n\nBefore the first edit, write down:\n\n- **What am I actually building?** One sentence. If you cannot state it, stop.\n- **What am I assuming?** List every assumption. If any is load-bearing and\n  unverified, flag `NEEDS_CONTEXT` and ask — do not guess.\n- **What does success look like?** Map each acceptance criterion to a concrete\n  test, check, or observable behaviour. If a criterion is not verifiable,\n  propose a sharpening via the `revision` skill (automated-trace subsection), not a vague attempt.\n\nRefusing to produce code is allowed. A task with unknown scope is a spec bug,\nnot a coding task.\n\n## 2. Simplicity First\n\nThe correct amount of code is the minimum that meets the acceptance criteria.\n\n- No speculative features. No abstraction layer \"in case we need it.\"\n- No new dependencies unless the task requires one and no existing dep fits.\n- No \"while I'm in here\" refactors. Surface them as separate kits.\n- Duplication is not always wrong. Three similar lines usually beat a premature\n  abstraction with two configuration knobs.\n\nIf the diff is larger than the acceptance criteria seem to demand, explain why\nin the commit body. If you cannot, trim the diff.\n\n## 3. Surgical Changes\n\nEvery line in the diff must trace back to an acceptance criterion. Touching\ncode outside the task's owned files is justified only when a requirement forces\nit. Examples of violations:\n\n- Fixing a formatter warning in an unrelated file.\n- Renaming a helper \"to match new convention.\"\n- Reordering imports, docstrings, whitespace.\n- Tightening a type signature the task did not ask about.\n\nIf you see a real bug in adjacent code, log it to `.cavekit\u002Fhistory\u002Fbackprop-log.md`\nas a candidate kit item and keep it out of this task's diff.\n\n## 4. Goal-Driven Execution\n\nTransform vague tasks into verifiable success criteria before execution.\n\n- A task that cannot be verified is not a task — escalate it.\n- The verification plan must be concrete: exact commands, exact assertions,\n  exact files to inspect. \"Make sure it works\" is not a plan.\n- After implementation, run the verification plan. Report the output.\n\n## Role-specific enforcement\n\n- **task-builder** — must produce, alongside code, a Verification Report listing\n  each AC, the verification step, and the observed result.\n- **reviewer** — must refuse to advance to Pass 2 (code quality) if Pass 1 finds\n  any of: undeclared assumptions, diff lines unjustified by an AC, out-of-scope\n  edits, or unreachable verification steps.\n- **planner** — must reject kits that contain un-testable ACs. They are spec\n  bugs and block planning.\n- **inspector** — must flag completed tasks whose verification logs are missing\n  or hand-waved.\n\n## When you are tempted to break a rule\n\nYou are probably over-confident about a shortcut that will cost more than the\ndelay of asking. Stop and note the tension in the commit body or in the build\nlog so the reviewer can judge.\n",{"name":5,"description":7},"main","2b63f856ce19c2a797929cacabd9e0fc936812fa","plugins\u002FJuliusBrussee\u002Fblueprint\u002Fskills\u002Fkarpathy-guardrails",468,"https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins","https:\u002F\u002Fhol.org\u002Fplugins\u002Fbest-codex-plugins","https:\u002F\u002Fgithub.com\u002Fhashgraph-online","https:\u002F\u002Fhol.org","Open-source standards and developer tools for interoperable AI agents: discovery, identity, communication, payments, privacy and security.","2026-11-11 02:22:54",{"source":8,"skillName":5,"commands":56,"prompt":66,"zipAvailable":67},[57,60,63],{"tool":58,"command":59},"npx","npx skills add hashgraph-online\u002Fawesome-codex-plugins --skill karpathy-guardrails",{"tool":61,"command":62},"bunx","bunx skills add hashgraph-online\u002Fawesome-codex-plugins --skill karpathy-guardrails",{"tool":64,"command":65},"pnpm","pnpm dlx skills add hashgraph-online\u002Fawesome-codex-plugins --skill karpathy-guardrails","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002FJuliusBrussee\u002Fblueprint\u002Fskills\u002Fkarpathy-guardrails\u002FSKILL.md",true,[],[70,95,120,154,174,188],{"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":40},"01M716A6ETNB9B1SQ3GXZC75VR","deja-search","deja-vu memory — search the user's past AI coding sessions with the deja CLI. Holds what they already settled and never wrote into the repo (naming rules, banned or already-tried tools, required checks before merge, preferred values, who owns what). Use before your first edit in a task, before you call a change done or ready to merge, before you recommend a tool, library, design or reviewer, before re-debugging an error that may already be solved, when they say 'didn't we fix this before', 'what did we decide about X' or 'I already have that', and before telling them something on this machine does not exist.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-search\u002FSKILL.md",[],[77,78,79,80,81,82,83,84,35,85,86,87,88,89,37,90,91,92,93,94],"search","deja search","error","ready","never","memory","re","decide","design","debugging","when","you","done","io","change","cli","merge","past",{"skillNo":96,"name":97,"nameZh":6,"description":98,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":99,"categories":100,"keywords":101,"publishedAt":40},"01M716A3J2W5SENSE9VGMXB375","deja-reflect","End-of-session reflection and memory curation with deja-vu. Review what was decided, what commands broke and what repaired them, and record durable project rules or decisions with deja remember so the next session starts with them.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fvshulcz\u002Fdeja-vu\u002Fskills\u002Fdeja-reflect\u002FSKILL.md",[],[102,103,104,105,82,106,107,108,109,110,111,112,113,90,114,115,116,117,118,119],"reflect","deja reflect","brew","remember","project rules","project","session","end","reflection","Rules","decisions","review","with","record","github","next","https","node",{"skillNo":121,"name":122,"nameZh":6,"description":123,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":124,"categories":125,"keywords":137,"publishedAt":40},"01M715T3RR4BAYSR57HPPTJ0PF","revid","Make, edit, export and post AI videos with Revid (revid.ai) through the Revid MCP server. Use whenever the user wants a short-form or long-form video made (from an idea, a script, a URL, an article, a PDF, a tweet, a product page, their own clips, or a song), wants to find what is going viral on TikTok, YouTube, Instagram or X before making something, wants to edit or re-export an existing Revid project, generate a standalone AI image or clip, make a series episode, or publish and schedule a finished video to connected social accounts. Covers setup (hosted MCP with OAuth, or an API key), the 36 tools, credits, polling and the rules for publishing. Routes to the more specific revid-* skills when one fits.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Ftmakersas\u002Frevid-skills\u002Fskills\u002Frevid\u002FSKILL.md",[126,131,134],{"categoryNo":127,"categoryName":128,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000010","后端与接口","SCENARIO","使用场景",{"categoryNo":132,"categoryName":133,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000016","图像与多媒体",{"categoryNo":135,"categoryName":136,"taxonomyType":129,"taxonomyTypeDesc":130},"01JBSEEDCAT000000000000023","Skill与智能体开发",[122,138,139,140,141,142,107,143,83,144,145,146,147,87,148,149,150,151,152,153],"Video","short-form","standalone","PDF","viral","tiktok","short","social","MCP","song","URL","image","mcp server","schedule","api","publish",{"skillNo":155,"name":156,"nameZh":6,"description":157,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":158,"categories":159,"keywords":160,"publishedAt":40},"01M71569ZM2X3H5RBFK8PVKMK2","review-study","Review a QuestionPunk survey before launch for wording, answer choices, branching, and missing settings. Use for check my survey or review this draft before publishing.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Freview-study\u002FSKILL.md",[],[161,162,113,25,163,164,165,166,167,168,169,170,171,172,173],"study","review study","answer","missing","survey","publishing","use","check","draft","settings","launch","my","branching",{"skillNo":175,"name":176,"nameZh":6,"description":177,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":178,"categories":179,"keywords":180,"publishedAt":40},"01M71567GV0C35YQBY0DPGR86B","manage-study","Find QuestionPunk studies, check status, publish an explicitly requested draft, or retrieve its participant link. Use for find my survey, publish this study, or how many people responded.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fmanage-study\u002FSKILL.md",[],[181,161,182,25,165,167,183,168,169,153,184,185,186,187,172],"manage study","manage","people","how","find","link","status",{"skillNo":189,"name":190,"nameZh":6,"description":191,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":192,"categories":193,"keywords":194,"publishedAt":40},"01M71565A8TCA01AZQZRGRYMFZ","create-study","Create a saved QuestionPunk survey draft from a research brief or edit specified questions. Use for requests such as make a customer feedback survey or shorten my onboarding study.","https:\u002F\u002Fgithub.com\u002Fhashgraph-online\u002Fawesome-codex-plugins\u002Fblob\u002Fmain\u002Fplugins\u002Fquestionpunk\u002Fcodex-plugin\u002Fskills\u002Fcreate-study\u002FSKILL.md",[],[195,161,196,197,198,199,25,200,165,201,167,202,203,169,204,205,206,207,208,172],"create study","create","brief","as","feedback","customer feedback","research brief","shorten","saved","research","requests","customer","questions","edit"]