[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M79BN2GV1N2125BJSD3P4KHT":3,"skill-same:01M79BN2GV1N2125BJSD3P4KHT":81},{"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":27,"commitSha":28,"skillDirectory":29,"forksCount":14,"repositoryHtmlUrl":30,"repositoryHomepageUrl":31,"ownerHtmlUrl":32,"ownerBlogUrl":6,"ownerDescription":33,"updatedAt":34,"install":35,"resources":48},"01M79BN2GV1N2125BJSD3P4KHT","claude-code-docs",null,"Look up authoritative, up-to-date Claude Code documentation covering the CLI, settings, permissions, hooks, MCP, skills, plugins and marketplaces, subagents, agent teams, workflows, worktrees, sandboxing, channels, routines, scheduled tasks, the desktop\u002Fweb\u002Fmobile\u002FSlack\u002FIDE surfaces, code review, the Agent SDK (Python and TypeScript), gateways, Bedrock\u002FVertex\u002FFoundry deployment, enterprise admin, and troubleshooting. Use when the user asks how a Claude Code feature works, what a config field does, how to set up hooks\u002FMCP\u002Fskills\u002Fsubagents\u002Fplugins, when troubleshooting a Claude Code error or unexpected behavior, or when you need to cite current official docs rather than rely on training-cutoff knowledge. Also triggers on Chinese requests such as 查一下 Claude Code 文档、Claude Code 官方文档怎么说、Claude Code 的 hooks\u002FMCP\u002Fskills\u002F权限\u002F配置 怎么用.","MacHu-GWU\u002Fshsk_lesson_smith-project","MacHu-GWU","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F6800411?v=4","01M79BMG0V767YAMAMTG1ZQ7FY","Sanhe Hu","01M79BMG1XDNB3Q6B6A60FBMEH",0,"AGPL-3.0","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: claude-code-docs\ndescription: Look up authoritative, up-to-date Claude Code documentation covering the CLI, settings, permissions, hooks, MCP, skills, plugins and marketplaces, subagents, agent teams, workflows, worktrees, sandboxing, channels, routines, scheduled tasks, the desktop\u002Fweb\u002Fmobile\u002FSlack\u002FIDE surfaces, code review, the Agent SDK (Python and TypeScript), gateways, Bedrock\u002FVertex\u002FFoundry deployment, enterprise admin, and troubleshooting. Use when the user asks how a Claude Code feature works, what a config field does, how to set up hooks\u002FMCP\u002Fskills\u002Fsubagents\u002Fplugins, when troubleshooting a Claude Code error or unexpected behavior, or when you need to cite current official docs rather than rely on training-cutoff knowledge. Also triggers on Chinese requests such as 查一下 Claude Code 文档、Claude Code 官方文档怎么说、Claude Code 的 hooks\u002FMCP\u002Fskills\u002F权限\u002F配置 怎么用.\nargument-hint: [topic or doc title]\nallowed-tools: WebFetch\n---\n\n# Claude Code Docs\n\nAnswers Claude Code questions from the official docs on demand: WebFetches the `llms.txt`\nindex (a flat list of every page, with a real description per entry), picks the pages that\nmatch, and fetches those as raw markdown. Always prefer this skill over recalling docs from\nmemory — the docs change faster than training data.\n\nIf the user passed an argument (`$ARGUMENTS`), treat it as the topic to look up. Otherwise\ninfer the topic from the conversation.\n\n## When to use this skill\n\nUse it whenever the question is about Claude Code itself or anything in its docs scope:\n\n- **CLI and session surface** — slash commands, CLI flags, interactive mode, keybindings,\n  status line, output styles, fullscreen, voice dictation, sessions, checkpointing, worktrees\n- **Configuration** — settings, environment variables, permissions and permission modes,\n  memory\u002FCLAUDE.md, model config, the `.claude` directory, server-managed settings\n- **Extensibility** — hooks, MCP, skills, plugins, plugin marketplaces, subagents, agent\n  teams, dynamic workflows, channels, routines, scheduled tasks, computer use\n- **Agent SDK** (Python and TypeScript) — agent loop, sessions, streaming, custom tools,\n  hooks, permissions, structured outputs, tool search, hosting, observability\n- **Surfaces and integrations** — desktop app, web, mobile, Slack, Chrome, VS Code, JetBrains,\n  devcontainers, GitHub Actions, GitLab CI\u002FCD\n- **Enterprise and deployment** — admin setup, Bedrock, Google Cloud, Microsoft Foundry, LLM\n  gateways, Claude apps gateway, network config, analytics, costs, security, data usage\n- **Reference and troubleshooting** — tools reference, error reference, glossary, changelog,\n  the weekly `What's new` pages, troubleshooting\n\nOut of scope: the **Anthropic API \u002F Anthropic SDK** itself (Messages API, pricing, model ids).\nUse the `claude-api` skill for that.\n\n## How this site works\n\nMeasured 2026-07-30; full fact sheet in [references\u002Fmechanism.md](references\u002Fmechanism.md).\n\n- **Index**: `https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms.txt` — 38,847 B (~9,700 tokens), 174 entries,\n  100% with prose descriptions. One flat `## Docs` section; there is no sub-section routing to\n  lean on, so the descriptions are the only triage signal.\n- **Content**: every index URL already ends in `.md` and serves raw markdown as-is. Measured\n  `hooks.md` = 242,078 B of markdown vs 2,423,751 B for the HTML page — ~90% cheaper. There is\n  no separate `.md` twin to construct; use the URL exactly as the index gives it.\n- **Page sizes vary by 25×**: measured 10,628 B (`troubleshooting.md`) to 272,484 B\n  (`settings.md`); `hooks.md` 242,078 B, `cli-reference.md` 104,140 B, `mcp.md` 80,866 B,\n  `overview.md` 16,445 B. Ask WebFetch a pointed question — a vague prompt on a 270 KB page\n  wastes most of the fetch.\n- **Gotcha — `llms-full.txt`**: `https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms-full.txt` is 6,556,407 B\n  (~1.64M tokens). It is a full-text dump, not an index. Never fetch it.\n\n## Procedure\n\n### 1. Read the index\n\n```\nWebFetch url=https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms.txt\n        prompt=\"Return the raw markdown. I need every `- [Title](URL): description` line unmodified.\"\n```\n\nEach entry is `- [Title](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002F\u003Cslug>.md): description`.\n\nDon't skip this step, even if you think you remember the right URL. Slugs get renamed and\npages get added weekly; the index is the source of truth.\n\n### 2. Pick the right page(s)\n\nMatch against the **description** (the text after the colon), not just the title. Then:\n\n- Pick **1–3 pages per batch**, not more. The index is for triage, not bulk loading.\n- One specific feature (\"how do hooks work?\") → one page.\n- Cross-concept question (\"how do skills relate to subagents?\") → fetch each relevant page.\n- Note that some topics live in two places — e.g. `hooks.md` (reference) vs `hooks-guide.md`\n  (how-to) vs `agent-sdk\u002Fhooks.md` (SDK). Pick the one matching what was actually asked.\n\n### 3. If nothing in the index looks like a match\n\nDo **not** conclude the topic is undocumented. Escalate, in order:\n\n1. **Widen with synonyms.** The docs' word is often not the user's. Measured on this index:\n   \"resume\" appears in **zero** titles but in five descriptions (it lives under\n   `sessions.md`, *Manage sessions*). Same shape: \"auto-approve\" → *permission modes*;\n   \"parallel checkouts\" → *worktrees*; \"cron\" → *scheduled tasks* \u002F *routines*.\n2. **Translate a non-English query into English first.** This index is English-only. Measured:\n   the Chinese query `钩子` scores **0** matches while `hook` scores 17 — a non-English miss\n   says nothing about coverage. Never report \"not documented\" off a non-English search.\n3. **Ask \"what recently changed?\"** — `whats-new\u002Findex.md` plus the 18 weekly\n   `whats-new\u002F2026-wNN.md` pages cover new features that older, stabler pages don't mention yet.\n4. **Only then** say it is not in the docs, and state what you searched.\n\n### 4. Fetch the batch\n\nFor each chosen URL, use it verbatim from the index (it already ends in `.md`):\n\n```\nWebFetch url=\u003CURL from index>\n        prompt=\"\u003Ca question that captures what the user actually needs, not 'summarize this page'>\"\n```\n\n### 5. Evaluate, then loop or answer\n\nAfter each batch, judge whether the fetched pages actually answer the question:\n\n- **Enough** → answer, grounded in the fetched content. Cite the doc page (title + URL) when\n  stating non-obvious facts so the user can verify.\n- **Not enough** (the answer lives on a page you haven't read, or a fetched page pointed to\n  another) → go back to step 2, pick the next 1–3 pages, and fetch again.\n- Keep looping to a **default cap of 9 pages** total across all batches.\n- **Still not enough at 9 pages** → stop. Tell the user what you've read, what's still missing,\n  and ask whether to keep going. Don't silently blow past the cap or pad the answer with guesses.\n\n## Context budget\n\n| Step | Cost | Notes |\n| :--- | :--- | :--- |\n| index | ~9,700 tok | once per question; re-use it across batches in the same turn |\n| page | ~3k–68k tok raw | 1–3 per batch; WebFetch reduces this before it reaches context |\n\nTypical question: index + 1–2 pages. Re-read the index only if the conversation moves to an\nunrelated topic — not between batches.\n\n## Rules\n\n- **Never invent a doc URL.** If a page isn't in the index, it does not exist — say so instead\n  of fabricating a slug. A 404 on an index URL means the index is stale; re-run\n  `\u002Fdocs-skill-builder check` on this skill.\n- **Never fetch `llms-full.txt`** (6.5 MB \u002F ~1.64M tokens).\n- **Loop in small batches, cap at 9 pages.** Fetch 1–3, check if that's enough, fetch more only\n  if it isn't. Don't grind through the index or fabricate the gap.\n- **Stay in scope.** This skill covers `code.claude.com\u002Fdocs\u002F*` only. For the Anthropic API use\n  `claude-api`.\n- **Pass through what the docs say.** Don't merge aggressively with prior knowledge — the user\n  wants current authoritative behavior, not a synthesis.\n","# Claude Code Docs\n\nAnswers Claude Code questions from the official docs on demand: WebFetches the `llms.txt`\nindex (a flat list of every page, with a real description per entry), picks the pages that\nmatch, and fetches those as raw markdown. Always prefer this skill over recalling docs from\nmemory — the docs change faster than training data.\n\nIf the user passed an argument (`$ARGUMENTS`), treat it as the topic to look up. Otherwise\ninfer the topic from the conversation.\n\n## When to use this skill\n\nUse it whenever the question is about Claude Code itself or anything in its docs scope:\n\n- **CLI and session surface** — slash commands, CLI flags, interactive mode, keybindings,\n  status line, output styles, fullscreen, voice dictation, sessions, checkpointing, worktrees\n- **Configuration** — settings, environment variables, permissions and permission modes,\n  memory\u002FCLAUDE.md, model config, the `.claude` directory, server-managed settings\n- **Extensibility** — hooks, MCP, skills, plugins, plugin marketplaces, subagents, agent\n  teams, dynamic workflows, channels, routines, scheduled tasks, computer use\n- **Agent SDK** (Python and TypeScript) — agent loop, sessions, streaming, custom tools,\n  hooks, permissions, structured outputs, tool search, hosting, observability\n- **Surfaces and integrations** — desktop app, web, mobile, Slack, Chrome, VS Code, JetBrains,\n  devcontainers, GitHub Actions, GitLab CI\u002FCD\n- **Enterprise and deployment** — admin setup, Bedrock, Google Cloud, Microsoft Foundry, LLM\n  gateways, Claude apps gateway, network config, analytics, costs, security, data usage\n- **Reference and troubleshooting** — tools reference, error reference, glossary, changelog,\n  the weekly `What's new` pages, troubleshooting\n\nOut of scope: the **Anthropic API \u002F Anthropic SDK** itself (Messages API, pricing, model ids).\nUse the `claude-api` skill for that.\n\n## How this site works\n\nMeasured 2026-07-30; full fact sheet in [references\u002Fmechanism.md](references\u002Fmechanism.md).\n\n- **Index**: `https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms.txt` — 38,847 B (~9,700 tokens), 174 entries,\n  100% with prose descriptions. One flat `## Docs` section; there is no sub-section routing to\n  lean on, so the descriptions are the only triage signal.\n- **Content**: every index URL already ends in `.md` and serves raw markdown as-is. Measured\n  `hooks.md` = 242,078 B of markdown vs 2,423,751 B for the HTML page — ~90% cheaper. There is\n  no separate `.md` twin to construct; use the URL exactly as the index gives it.\n- **Page sizes vary by 25×**: measured 10,628 B (`troubleshooting.md`) to 272,484 B\n  (`settings.md`); `hooks.md` 242,078 B, `cli-reference.md` 104,140 B, `mcp.md` 80,866 B,\n  `overview.md` 16,445 B. Ask WebFetch a pointed question — a vague prompt on a 270 KB page\n  wastes most of the fetch.\n- **Gotcha — `llms-full.txt`**: `https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms-full.txt` is 6,556,407 B\n  (~1.64M tokens). It is a full-text dump, not an index. Never fetch it.\n\n## Procedure\n\n### 1. Read the index\n\n```\nWebFetch url=https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fllms.txt\n        prompt=\"Return the raw markdown. I need every `- [Title](URL): description` line unmodified.\"\n```\n\nEach entry is `- [Title](https:\u002F\u002Fcode.claude.com\u002Fdocs\u002Fen\u002F\u003Cslug>.md): description`.\n\nDon't skip this step, even if you think you remember the right URL. Slugs get renamed and\npages get added weekly; the index is the source of truth.\n\n### 2. Pick the right page(s)\n\nMatch against the **description** (the text after the colon), not just the title. Then:\n\n- Pick **1–3 pages per batch**, not more. The index is for triage, not bulk loading.\n- One specific feature (\"how do hooks work?\") → one page.\n- Cross-concept question (\"how do skills relate to subagents?\") → fetch each relevant page.\n- Note that some topics live in two places — e.g. `hooks.md` (reference) vs `hooks-guide.md`\n  (how-to) vs `agent-sdk\u002Fhooks.md` (SDK). Pick the one matching what was actually asked.\n\n### 3. If nothing in the index looks like a match\n\nDo **not** conclude the topic is undocumented. Escalate, in order:\n\n1. **Widen with synonyms.** The docs' word is often not the user's. Measured on this index:\n   \"resume\" appears in **zero** titles but in five descriptions (it lives under\n   `sessions.md`, *Manage sessions*). Same shape: \"auto-approve\" → *permission modes*;\n   \"parallel checkouts\" → *worktrees*; \"cron\" → *scheduled tasks* \u002F *routines*.\n2. **Translate a non-English query into English first.** This index is English-only. Measured:\n   the Chinese query `钩子` scores **0** matches while `hook` scores 17 — a non-English miss\n   says nothing about coverage. Never report \"not documented\" off a non-English search.\n3. **Ask \"what recently changed?\"** — `whats-new\u002Findex.md` plus the 18 weekly\n   `whats-new\u002F2026-wNN.md` pages cover new features that older, stabler pages don't mention yet.\n4. **Only then** say it is not in the docs, and state what you searched.\n\n### 4. Fetch the batch\n\nFor each chosen URL, use it verbatim from the index (it already ends in `.md`):\n\n```\nWebFetch url=\u003CURL from index>\n        prompt=\"\u003Ca question that captures what the user actually needs, not 'summarize this page'>\"\n```\n\n### 5. Evaluate, then loop or answer\n\nAfter each batch, judge whether the fetched pages actually answer the question:\n\n- **Enough** → answer, grounded in the fetched content. Cite the doc page (title + URL) when\n  stating non-obvious facts so the user can verify.\n- **Not enough** (the answer lives on a page you haven't read, or a fetched page pointed to\n  another) → go back to step 2, pick the next 1–3 pages, and fetch again.\n- Keep looping to a **default cap of 9 pages** total across all batches.\n- **Still not enough at 9 pages** → stop. Tell the user what you've read, what's still missing,\n  and ask whether to keep going. Don't silently blow past the cap or pad the answer with guesses.\n\n## Context budget\n\n| Step | Cost | Notes |\n| :--- | :--- | :--- |\n| index | ~9,700 tok | once per question; re-use it across batches in the same turn |\n| page | ~3k–68k tok raw | 1–3 per batch; WebFetch reduces this before it reaches context |\n\nTypical question: index + 1–2 pages. Re-read the index only if the conversation moves to an\nunrelated topic — not between batches.\n\n## Rules\n\n- **Never invent a doc URL.** If a page isn't in the index, it does not exist — say so instead\n  of fabricating a slug. A 404 on an index URL means the index is stale; re-run\n  `\u002Fdocs-skill-builder check` on this skill.\n- **Never fetch `llms-full.txt`** (6.5 MB \u002F ~1.64M tokens).\n- **Loop in small batches, cap at 9 pages.** Fetch 1–3, check if that's enough, fetch more only\n  if it isn't. Don't grind through the index or fabricate the gap.\n- **Stay in scope.** This skill covers `code.claude.com\u002Fdocs\u002F*` only. For the Anthropic API use\n  `claude-api`.\n- **Pass through what the docs say.** Don't merge aggressively with prior knowledge — the user\n  wants current authoritative behavior, not a synthesis.\n",{"name":5,"description":7,"allowed-tools":24,"argument-hint":25},"WebFetch",[26],"topic or doc title","main","ef0f8eefce29dea12070e85b2da80f4be0b9d4fb",".claude\u002Fskills\u002Fclaude-code-docs","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project","https:\u002F\u002Fshsk-lesson-smith.readthedocs.io\u002F","https:\u002F\u002Fgithub.com\u002FMacHu-GWU","Founder mindset, engineer’s hands. Customers first, product second. Forward-deployed AI architect · ex-AWS · 200+ OSS libs · 4K+ commits\u002Fyr","2026-10-01 01:47:58",{"source":8,"skillName":5,"commands":36,"prompt":46,"zipAvailable":47},[37,40,43],{"tool":38,"command":39},"npx","npx skills add MacHu-GWU\u002Fshsk_lesson_smith-project --skill claude-code-docs",{"tool":41,"command":42},"bunx","bunx skills add MacHu-GWU\u002Fshsk_lesson_smith-project --skill claude-code-docs",{"tool":44,"command":45},"pnpm","pnpm dlx skills add MacHu-GWU\u002Fshsk_lesson_smith-project --skill claude-code-docs","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FSKILL.md",true,[49,55,59,63,67,71,75],{"path":50,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":53,"sourceHtmlUrl":54},"CHANGELOG.md","REFERENCE","参考文档",3436,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FCHANGELOG.md",{"path":56,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":57,"sourceHtmlUrl":58},"README-cn.md",10584,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FREADME-cn.md",{"path":60,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":61,"sourceHtmlUrl":62},"README.md",10072,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FREADME.md",{"path":64,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":65,"sourceHtmlUrl":66},"references\u002Fmechanism-cn.md",4392,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002Freferences\u002Fmechanism-cn.md",{"path":68,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":69,"sourceHtmlUrl":70},"references\u002Fmechanism.md",4011,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002Freferences\u002Fmechanism.md",{"path":72,"resourceType":51,"resourceTypeDesc":52,"sizeBytes":73,"sourceHtmlUrl":74},"SKILL-cn.md",8306,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FSKILL-cn.md",{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"VERSION","OTHER","其他",6,"https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fclaude-code-docs\u002FVERSION",[82,89,96,101,106,112],{"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},"01M79BVJQ839NXWYW9YG7PSVNF","upskill-quiz-cn","这门课的面试式自测. 一道道问下去, 逼你说清为什么, 而不只是是什么.","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_upskill_repo\u002F.claude\u002Fskills\u002Fupskill-quiz-cn\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},"01M79BVF959YEX1W6B1AR7AAX0","upskill-learn-cn","这门课的领路人. 要地图, 要拆开某一处, 或者卡住了出不来, 都找它.","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_upskill_repo\u002F.claude\u002Fskills\u002Fupskill-learn-cn\u002FSKILL.md",[],[],{"skillNo":97,"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":98,"categories":99,"keywords":100,"publishedAt":19},"01M79BVDFCA7GYEQ6Q4XV1RWVB","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_upskill_repo\u002F.agents\u002Fskills\u002Fupskill-quiz-cn\u002FSKILL.md",[],[],{"skillNo":102,"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":103,"categories":104,"keywords":105,"publishedAt":19},"01M79BVC28KMZ3APP4N3WQ0QQ2","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_upskill_repo\u002F.agents\u002Fskills\u002Fupskill-learn-cn\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"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":109,"categories":110,"keywords":111,"publishedAt":19},"01M79BV8AZ8SSB4R4ZKCNHZW8N","showcase-quiz-cn","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_showcase_repo\u002F.claude\u002Fskills\u002Fshowcase-quiz-cn\u002FSKILL.md",[],[],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":19},"01M79BV4TQSJTJWW9B36K11C7Y","showcase-publish-cn","把这个 repo 变成能放上你自己 GitHub 的作品. 抹掉教学痕迹, 再回头查一遍.","https:\u002F\u002Fgithub.com\u002FMacHu-GWU\u002Fshsk_lesson_smith-project\u002Fblob\u002Fmain\u002Ftests\u002Fgood_showcase_repo\u002F.claude\u002Fskills\u002Fshowcase-publish-cn\u002FSKILL.md",[],[]]