[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6C2716CPRC4FAQGKDYRG91B":3,"skill-same:01M6C2716CPRC4FAQGKDYRG91B":56},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":30,"linkOnly":31,"markdownContent":32,"markdownBody":33,"frontMatter":34,"refName":35,"commitSha":36,"skillDirectory":37,"forksCount":38,"repositoryHtmlUrl":39,"repositoryHomepageUrl":6,"ownerHtmlUrl":40,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":41,"install":42,"resources":55},"01M6C2716CPRC4FAQGKDYRG91B","karpathy-wiki",null,"Build, ingest, query, and lint an LLM wiki using the Karpathy knowledge-graph pattern.","jleechanorg\u002Fjleechan-skills","jleechanorg","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F221404966?v=4","01M6C1VJ99XG950X1DWPEBCBGX","01M6C1VJDK5YVDNT3362RT5WSB",6,"MIT","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fkarpathy-wiki\u002FSKILL.md",[],[18,19,20,21,22,23,24,25,26,27,28,29],"wiki","karpathy","karpathy wiki","lint","graph","knowledge-graph","knowledge","llm wiki","ingest","build","query","pattern","2026-11-10 10:12:30",false,"---\nname: karpathy-wiki\ndescription: Build, ingest, query, and lint an LLM wiki using the Karpathy knowledge-graph pattern.\n---\n\n# karpathy-wiki Skill\n\n## When to use\nWhen working with an LLM wiki (Karpathy pattern) - ingesting sources, querying, linting, or building the knowledge graph.\n\n## What it does\nEnforces the exact Karpathy LLM Wiki pattern from https:\u002F\u002Fgist.github.com\u002Fkarpathy\u002F442a6bf555914893e9891c11519de94f\n\n## Pattern Requirements\n\n### Directory Structure (MUST FOLLOW EXACTLY)\n```\nllm_wiki\u002F\n├── raw\u002F              # Immutable source documents (NEVER edit these)\n├── wiki\u002F             # LLM owns this layer entirely - single source of truth\n│   ├── index.md      # Catalog of ALL pages with curated one-line summaries\n│   ├── log.md        # Append-only chronological record (format: ## [YYYY-MM-DD] operation | title)\n│   ├── overview.md  # Living synthesis across all sources\n│   ├── sources\u002F      # Source summary pages (kebab-case.md)\n│   ├── entities\u002F    # Entity pages (TitleCase.md - people, companies, projects)\n│   ├── concepts\u002F     # Concept pages (TitleCase.md - ideas, methods, theories)\n│   └── syntheses\u002F   # Saved query answers\n└── (no root-level wiki files - ALL must be under wiki\u002F)\n```\n\n### Frontmatter (REQUIRED on every page)\n```yaml\n---\ntitle: \"Page Title\"\ntype: source | entity | concept | synthesis\ntags: []\nsources: []      # list of source slugs that inform this page\nlast_updated: YYYY-MM-DD\n---\n```\n\n### Page Naming\n- **Sources**: kebab-case (e.g., `json-display-bugs-analysis-report.md`)\n- **Entities**: TitleCase (e.g., `PR278.md`, `OpenAI.md`, `SamAltman.md`)\n- **Concepts**: TitleCase (e.g., `StructuredResponse.md`, `RAG.md`)\n\n### Wikilinks\nUse `[[PageName]]` for internal links - these become the knowledge graph edges.\n\n### Index Format\n```markdown\n# Wiki Index\n\n## Overview\n- [Overview](overview.md) — one-line synthesis\n\n## Sources\n- [Source Title](sources\u002Fslug.md) — CURATED one-line summary (NOT raw email subject)\n\n## Entities\n- [Entity Name](entities\u002FEntityName.md) — one-line description\n\n## Concepts\n- [Concept Name](concepts\u002FConceptName.md) — one-line description\n\n## Syntheses\n- [Analysis Title](syntheses\u002Fslug.md) — what question it answers\n```\n\n### Log Format\n```\n## [YYYY-MM-DD] operation | title\n\nKey claims: ...\n```\nOperations: `ingest`, `query`, `lint`, `graph`\n\n## Ingest Workflow\n1. Read source document from `raw\u002F`\n2. Read `wiki\u002Findex.md` and `wiki\u002Foverview.md` for context\n3. Write `wiki\u002Fsources\u002F\u003Cslug>.md` with Source Page Format\n4. Update `wiki\u002Findex.md` — add entry under correct section\n5. Update `wiki\u002Foverview.md` — revise synthesis\n6. Create\u002Fupdate entity pages for key people\u002Fcompanies\u002Fprojects\n7. Create\u002Fupdate concept pages for key ideas\u002Fmethods\n8. Flag any contradictions\n9. Append to `wiki\u002Flog.md`\n\n### Source Page Format\n```markdown\n---\ntitle: \"Source Title\"\ntype: source\ntags: []\ndate: YYYY-MM-DD\nsource_file: raw\u002F...\nsources: []\nlast_updated: YYYY-MM-DD\n---\n\n## Summary\n2–4 sentence summary.\n\n## Key Claims\n- Claim 1\n- Claim 2\n\n## Key Quotes\n> \"Quote here\" — context\n\n## Connections\n- [[EntityName]] — how they relate\n- [[ConceptName]] — how it connects\n\n## Contradictions\n- Contradicts [[OtherPage]] on: ...\n```\n\n## Query Workflow\n1. Read `wiki\u002Findex.md` to find relevant pages\n2. Read those pages\n3. Synthesize answer with `[[PageName]]` citations\n4. Offer to save as `wiki\u002Fsyntheses\u002F\u003Cslug>.md`\n\n## Lint Workflow\nCheck for:\n- **Orphans**: pages with no inbound `[[links]]`\n- **Broken links**: `[[WikiLinks]]` to non-existent pages\n- **Contradictions**: conflicting claims across pages\n- **Stale summaries**: pages not updated after newer sources\n- **Missing entities**: entities mentioned 3+ times without own page\n- **Missing concepts**: concepts discussed without own page\n- **Data gaps**: questions wiki can't answer\n\n## Graph Workflow\n1. Grep all `[[wikilinks]]` → deterministic `EXTRACTED` edges\n2. Infer implicit relationships → `INFERRED` edges with confidence\n3. Run Louvain community detection\n4. Output `graph\u002Fgraph.json` + `graph\u002Fgraph.html`\n\n## Common Issues to Fix\n\n### Dual Structure Problem\nIf wiki exists in both root AND `wiki\u002F` subdirectory:\n- Move all wiki content to `wiki\u002F` subdirectory only\n- Delete root-level `index.md`, `log.md`, `overview.md` if duplicates exist\n- Update ingest.py to write to `wiki\u002F` subdirectory\n\n### Index Quality Problem\nIf index contains raw email subjects instead of curated summaries:\n- Re-ingest with instruction to create proper summaries\n- Edit index entries to be content-oriented, not raw titles\n\n### Entity\u002FConcept Gap\nIf few entity\u002Fconcept pages created:\n- Force extraction: prompt must explicitly create entity pages for EVERY person\u002Fcompany\u002Fproject mentioned\n- Force extraction: prompt must explicitly create concept pages for EVERY idea\u002Fmethod\u002Ftheory\n- Track entity\u002Fconcept ratio - should be roughly 1:1 with sources or higher\n","# karpathy-wiki Skill\n\n## When to use\nWhen working with an LLM wiki (Karpathy pattern) - ingesting sources, querying, linting, or building the knowledge graph.\n\n## What it does\nEnforces the exact Karpathy LLM Wiki pattern from https:\u002F\u002Fgist.github.com\u002Fkarpathy\u002F442a6bf555914893e9891c11519de94f\n\n## Pattern Requirements\n\n### Directory Structure (MUST FOLLOW EXACTLY)\n```\nllm_wiki\u002F\n├── raw\u002F              # Immutable source documents (NEVER edit these)\n├── wiki\u002F             # LLM owns this layer entirely - single source of truth\n│   ├── index.md      # Catalog of ALL pages with curated one-line summaries\n│   ├── log.md        # Append-only chronological record (format: ## [YYYY-MM-DD] operation | title)\n│   ├── overview.md  # Living synthesis across all sources\n│   ├── sources\u002F      # Source summary pages (kebab-case.md)\n│   ├── entities\u002F    # Entity pages (TitleCase.md - people, companies, projects)\n│   ├── concepts\u002F     # Concept pages (TitleCase.md - ideas, methods, theories)\n│   └── syntheses\u002F   # Saved query answers\n└── (no root-level wiki files - ALL must be under wiki\u002F)\n```\n\n### Frontmatter (REQUIRED on every page)\n```yaml\n---\ntitle: \"Page Title\"\ntype: source | entity | concept | synthesis\ntags: []\nsources: []      # list of source slugs that inform this page\nlast_updated: YYYY-MM-DD\n---\n```\n\n### Page Naming\n- **Sources**: kebab-case (e.g., `json-display-bugs-analysis-report.md`)\n- **Entities**: TitleCase (e.g., `PR278.md`, `OpenAI.md`, `SamAltman.md`)\n- **Concepts**: TitleCase (e.g., `StructuredResponse.md`, `RAG.md`)\n\n### Wikilinks\nUse `[[PageName]]` for internal links - these become the knowledge graph edges.\n\n### Index Format\n```markdown\n# Wiki Index\n\n## Overview\n- [Overview](overview.md) — one-line synthesis\n\n## Sources\n- [Source Title](sources\u002Fslug.md) — CURATED one-line summary (NOT raw email subject)\n\n## Entities\n- [Entity Name](entities\u002FEntityName.md) — one-line description\n\n## Concepts\n- [Concept Name](concepts\u002FConceptName.md) — one-line description\n\n## Syntheses\n- [Analysis Title](syntheses\u002Fslug.md) — what question it answers\n```\n\n### Log Format\n```\n## [YYYY-MM-DD] operation | title\n\nKey claims: ...\n```\nOperations: `ingest`, `query`, `lint`, `graph`\n\n## Ingest Workflow\n1. Read source document from `raw\u002F`\n2. Read `wiki\u002Findex.md` and `wiki\u002Foverview.md` for context\n3. Write `wiki\u002Fsources\u002F\u003Cslug>.md` with Source Page Format\n4. Update `wiki\u002Findex.md` — add entry under correct section\n5. Update `wiki\u002Foverview.md` — revise synthesis\n6. Create\u002Fupdate entity pages for key people\u002Fcompanies\u002Fprojects\n7. Create\u002Fupdate concept pages for key ideas\u002Fmethods\n8. Flag any contradictions\n9. Append to `wiki\u002Flog.md`\n\n### Source Page Format\n```markdown\n---\ntitle: \"Source Title\"\ntype: source\ntags: []\ndate: YYYY-MM-DD\nsource_file: raw\u002F...\nsources: []\nlast_updated: YYYY-MM-DD\n---\n\n## Summary\n2–4 sentence summary.\n\n## Key Claims\n- Claim 1\n- Claim 2\n\n## Key Quotes\n> \"Quote here\" — context\n\n## Connections\n- [[EntityName]] — how they relate\n- [[ConceptName]] — how it connects\n\n## Contradictions\n- Contradicts [[OtherPage]] on: ...\n```\n\n## Query Workflow\n1. Read `wiki\u002Findex.md` to find relevant pages\n2. Read those pages\n3. Synthesize answer with `[[PageName]]` citations\n4. Offer to save as `wiki\u002Fsyntheses\u002F\u003Cslug>.md`\n\n## Lint Workflow\nCheck for:\n- **Orphans**: pages with no inbound `[[links]]`\n- **Broken links**: `[[WikiLinks]]` to non-existent pages\n- **Contradictions**: conflicting claims across pages\n- **Stale summaries**: pages not updated after newer sources\n- **Missing entities**: entities mentioned 3+ times without own page\n- **Missing concepts**: concepts discussed without own page\n- **Data gaps**: questions wiki can't answer\n\n## Graph Workflow\n1. Grep all `[[wikilinks]]` → deterministic `EXTRACTED` edges\n2. Infer implicit relationships → `INFERRED` edges with confidence\n3. Run Louvain community detection\n4. Output `graph\u002Fgraph.json` + `graph\u002Fgraph.html`\n\n## Common Issues to Fix\n\n### Dual Structure Problem\nIf wiki exists in both root AND `wiki\u002F` subdirectory:\n- Move all wiki content to `wiki\u002F` subdirectory only\n- Delete root-level `index.md`, `log.md`, `overview.md` if duplicates exist\n- Update ingest.py to write to `wiki\u002F` subdirectory\n\n### Index Quality Problem\nIf index contains raw email subjects instead of curated summaries:\n- Re-ingest with instruction to create proper summaries\n- Edit index entries to be content-oriented, not raw titles\n\n### Entity\u002FConcept Gap\nIf few entity\u002Fconcept pages created:\n- Force extraction: prompt must explicitly create entity pages for EVERY person\u002Fcompany\u002Fproject mentioned\n- Force extraction: prompt must explicitly create concept pages for EVERY idea\u002Fmethod\u002Ftheory\n- Track entity\u002Fconcept ratio - should be roughly 1:1 with sources or higher\n",{"name":5,"description":7},"main","42ccd02cbf5e8e47839dc3eab6fc754dc6b339cf",".claude\u002Fskills\u002Fkarpathy-wiki",0,"https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills","https:\u002F\u002Fgithub.com\u002Fjleechanorg","2026-11-11 09:18:40",{"source":8,"skillName":5,"commands":43,"prompt":53,"zipAvailable":54},[44,47,50],{"tool":45,"command":46},"npx","npx skills add jleechanorg\u002Fjleechan-skills --skill karpathy-wiki",{"tool":48,"command":49},"bunx","bunx skills add jleechanorg\u002Fjleechan-skills --skill karpathy-wiki",{"tool":51,"command":52},"pnpm","pnpm dlx skills add jleechanorg\u002Fjleechan-skills --skill karpathy-wiki","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fkarpathy-wiki\u002FSKILL.md",true,[],[57,75,97,111,125,141],{"skillNo":58,"name":59,"nameZh":6,"description":60,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":61,"categories":62,"keywords":63,"publishedAt":30},"01M716Y5MTJ6GYJZ2ATXCJK8G4","instinct","Use when the user invokes \u002Finstinct or \u002Fin, or asks to \"message Instinct\", \"ask Instinct\", \"tell Instinct\", \"check Instinct\", or read replies from the configured peer assistant. iMessage first, WhatsApp Web as fallback.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Finstinct\u002FSKILL.md",[],[59,64,65,66,67,68,69,70,71,72,73,74],"user","when","replies","check","use","identity","whatsapp","gmail","read","message","ask",{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":86,"publishedAt":30},"01M716Y4HF7FVVMJSA2CCVN5VF","hr","Compatibility alias for history-resume; produce the same fresh-state, provenance-aware resume briefing.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhr\u002FSKILL.md",[81],{"categoryNo":82,"categoryName":83,"taxonomyType":84,"taxonomyTypeDesc":85},"01JBSEEDCAT000000000000038","人力资源","PROFESSION","适用职业",[77,87,88,89,90,91,92,93,94,95,96],"resume","compatibility","provenance","for","history","produce","alias","briefing","state","fresh",{"skillNo":98,"name":99,"nameZh":6,"description":100,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":101,"categories":102,"keywords":104,"publishedAt":30},"01M716Y1FFNFKFJYYKF8HQA6T8","history-resume","Resume briefing for the current repository branch or worktree using fresh state and admitted history sources.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fhistory-resume\u002FSKILL.md",[103],{"categoryNo":82,"categoryName":83,"taxonomyType":84,"taxonomyTypeDesc":85},[91,87,105,106,64,107,94,108,90,109,110,95,96],"history resume","branch","analysis","repository","worktree","sources",{"skillNo":112,"name":113,"nameZh":6,"description":114,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":115,"categories":116,"keywords":117,"publishedAt":30},"01M716XWXJ8EPHK4RH9T14TR85","grok","Alias for the canonical grokbot browser conversation workflow; \u002Fgrok talks to the Grok bot the same way \u002Fdot talks to Dot.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fgrok\u002FSKILL.md",[],[113,118,119,120,90,121,93,122,123,124],"dot","bot","workflow","canonical","conversation","browser","grokbot",{"skillNo":126,"name":127,"nameZh":6,"description":128,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":129,"categories":130,"keywords":131,"publishedAt":30},"01M70N9Q07DTE0EW5KFXMMMSJ0","codexa-bench","Compare `codexa` and normal `codex` on a fair split of comparable work by default; use `--strict` for duplicate runs of the exact same task.","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fcodexa-bench\u002FSKILL.md",[],[132,133,134,68,135,90,136,137,138,139,140],"codexa bench","work","split","default","duplicate","task","compare","normal","codex",{"skillNo":142,"name":143,"nameZh":6,"description":144,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":145,"licenseFileUrl":146,"sourceHtmlUrl":147,"categories":148,"keywords":160,"publishedAt":30},"01M6C24S2X60SP8BBFGAEKPEGW","frontend-design","Create distinctive, production-grade frontend interfaces with high design quality. Use this skill when the user asks to build web components, pages, artifacts, posters, or applications (examples include websites, landing pages, dashboards, React components, HTML\u002FCSS layouts, or when styling\u002Fbeautifying any web UI). Generates creative, polished code and UI design that avoids generic AI aesthetics.","Apache-2.0","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FLICENSE.txt","https:\u002F\u002Fgithub.com\u002Fjleechanorg\u002Fjleechan-skills\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Ffrontend-design\u002FSKILL.md",[149,154,157],{"categoryNo":150,"categoryName":151,"taxonomyType":152,"taxonomyTypeDesc":153},"01JBSEEDCAT000000000000009","前端开发","SCENARIO","使用场景",{"categoryNo":155,"categoryName":156,"taxonomyType":152,"taxonomyTypeDesc":153},"01JBSEEDCAT000000000000015","视觉设计",{"categoryNo":158,"categoryName":159,"taxonomyType":84,"taxonomyTypeDesc":85},"01JBSEEDCAT000000000000032","设计师",[161,162,163,68,164,165,166,64,167,168,169,170,171,172,173,174,175,176,177,178,179,180,181,182,183,184,65,185],"frontend","design","frontend design","generic","ui design","react","html","components","code","artifacts","ui","aesthetics","Production","web components","landing","applications","quality","web","css","create","examples","with","react components","web ui","pages"]