[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M71JM2T5FBPBQ855F3DSJ1FF":3,"skill-same:01M71JM2T5FBPBQ855F3DSJ1FF":82},{"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":23,"publishedAt":44,"linkOnly":45,"markdownContent":46,"markdownBody":47,"frontMatter":48,"refName":49,"commitSha":50,"skillDirectory":5,"forksCount":51,"repositoryHtmlUrl":52,"repositoryHomepageUrl":53,"ownerHtmlUrl":54,"ownerBlogUrl":55,"ownerDescription":56,"updatedAt":57,"install":58,"resources":71},"01M71JM2T5FBPBQ855F3DSJ1FF","obsidian-convert",null,"Convert source files (PDFs, DOCX, image-only PDFs) into Obsidian-flavored markdown notes that preserve the original content **verbatim** while applying proper Obsidian styling — YAML frontmatter with tags\u002Faliases, callouts, LaTeX math, footnotes, wikilinks, and tables. Use this skill whenever the user wants to convert, transcribe, extract, rebuild, \"turn into a note\", or \"make an Obsidian version\" of a PDF \u002F source document, even if they don't explicitly say \"Obsidian\". Also use it when the user mentions textbook chapters, lecture slides, exam papers, formula sheets, definition lists, or revision materials that need to become markdown notes in their vault. Triggers on phrases like \"convert this PDF\", \"make me a note from\", \"transcribe this into markdown\", \"rebuild this as markdown\", \"extract verbatim\", \"turn this into a study note\", or when the user points at a `.pdf`\u002F`.docx` in their vault and asks to do something with the contents.","caezium\u002Fskills","caezium","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F113233555?v=4","01M71JB77ZPBAWEK009Y2W8BP7","Henry Zhang","01M71JB792NVXCWSYD27HQ0964",3,"MIT","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fobsidian-convert\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000002","文档处理","SCENARIO","使用场景",[24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43],"convert","obsidian convert","obsidian","revision","at","markdown","math","me","notes","rebuild","image","when","exam","files","only","formula","papers","definition","PDF","docx","2026-11-10 10:12:30",false,"---\nname: obsidian-convert\ndescription: Convert source files (PDFs, DOCX, image-only PDFs) into Obsidian-flavored markdown notes that preserve the original content **verbatim** while applying proper Obsidian styling — YAML frontmatter with tags\u002Faliases, callouts, LaTeX math, footnotes, wikilinks, and tables. Use this skill whenever the user wants to convert, transcribe, extract, rebuild, \"turn into a note\", or \"make an Obsidian version\" of a PDF \u002F source document, even if they don't explicitly say \"Obsidian\". Also use it when the user mentions textbook chapters, lecture slides, exam papers, formula sheets, definition lists, or revision materials that need to become markdown notes in their vault. Triggers on phrases like \"convert this PDF\", \"make me a note from\", \"transcribe this into markdown\", \"rebuild this as markdown\", \"extract verbatim\", \"turn this into a study note\", or when the user points at a `.pdf`\u002F`.docx` in their vault and asks to do something with the contents.\n---\n\n# Obsidian Convert\n\nConvert source files into Obsidian-flavored markdown notes that preserve **all** original content verbatim while applying the user's Obsidian styling conventions.\n\n## Core principle: verbatim fidelity\n\nThe most important rule is that **no content gets lost or paraphrased**. Every numbered point, every exemplar answer, every footnote annotation, every formula, every quoted passage must appear in the output exactly as written in the source. The user is studying from these notes — if you summarize, you've made the note useless for revision.\n\nYou're allowed to **restructure** (split a wall of text into bullets, lift Q&A into callouts, turn comparisons into tables) but you're **not** allowed to **abbreviate**.\n\nThe only things you should strip are pure layout artifacts:\n- Page numbers (e.g., the standalone `1`, `2`, `3` at the bottom of each page)\n- Repeated page headers \u002F running headers (e.g., `吉屋教研组` reappearing at the top of every page)\n- Decorative whitespace between page breaks\n- PDF font-substitution artifacts (orphan ligature glyphs like `fi`, `fl` floating on their own lines — rejoin them into their words)\n\nIf you're ever unsure whether to keep something, keep it.\n\n## Workflow\n\n### 1. Identify the source type and extract text\n\nSource PDFs come in three flavors. Detect which one you have before deciding how to extract.\n\n**Text-extractable PDFs** (the common case — textbook chapters, exam papers, formula sheets typeset in TeX\u002FWord):\n```bash\nmkdir -p \u002Ftmp\u002Fpdf_extracts\npdftotext -layout \"\u003Csource.pdf>\" \u002Ftmp\u002Fpdf_extracts\u002F\u003Cname>.txt\nwc -l \u002Ftmp\u002Fpdf_extracts\u002F\u003Cname>.txt\n```\nIf the extract has more than ~50 lines and you can read text in it (not just garbage), it worked. Read the `.txt` instead of the PDF — bypasses the 32 MB API attachment limit and is much faster.\n\n**Image-only PDFs** (scans, photos, handwritten notes, slide decks with everything rendered as images): `pdftotext` returns \u003C 2 KB of mostly nothing. Fall back to `Read` with the `pages` parameter (max 20 per call):\n```\nRead(\"\u003Csource.pdf>\", pages=\"1-20\")\n```\nRead in chunks if the document is longer.\n\n**DOCX**: try `pandoc -t markdown \u003Cfile.docx>` if pandoc is installed, otherwise note that you couldn't extract and ask the user how to proceed.\n\n### 2. Plan the structure before writing\n\nBefore opening `Write`, skim the entire extracted text and decide:\n- What's the H1 title? (Usually the document's actual title.)\n- What are the major H2 sections? (Chapter sections, topic groupings, syllabus units.)\n- Where are the point lists, exemplar answers, Q&A blocks, tables, equations? Each needs different styling — see [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md).\n- What footnote markers appear? Map them to `[^1]`, `[^2]`… and decide where the definitions live (inline after the paragraph that references them, or all at the bottom).\n- Are there obvious related notes already in the vault to wikilink to? (e.g., a Chapter 3 note should link `[[Chapter 2]]` and `[[Chapter 4]]` for navigation.)\n\nQuick mental check: if you can't answer \"what's the H1 and what are the H2s\" after skimming, you don't understand the doc well enough yet.\n\n### 3. Write the note in one pass\n\nUse the `Write` tool with the full structured markdown. The full styling spec is in [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md) — read it before writing if you're unsure about a callout type or frontmatter convention. The most important conventions:\n\n**Frontmatter** (always):\n```yaml\n---\ndraft: false\ntags:\n  - \u003Csubject>      # e.g., history, physics, economics, chinese\n  - igcse          # or a-level, etc.\n  - \u003Ctopic-slug>   # e.g., chapter1, versailles\naliases:\n  - \u003CAlt name 1>\n  - \u003CChinese name if applicable>\n---\n```\n\n**Title line** (always): `# \u003Ctitle from source>`\n\n**Source attribution blockquote** (always, right after the H1):\n```markdown\n> Source PDF: `\u003Cfilename.pdf>`. \u003Cone-line description of what's in the doc>.\n```\n\n**Callout mapping** (see styling-rules.md for the full table):\n- Point lists → `> [!note]- Point List` (collapsed)\n- Exemplar \u002F model answers → `> [!example]- Exemplar Answer` (collapsed)\n- Definitions → `> [!def]`\n- Quoted source text \u002F poems \u002F passages to memorize → `> [!quote]`\n- Q&A discussion questions → `> [!question]`\n- Sidebar info, related-notes lists → `> [!info]`\n- Strategy tips, exam advice → `> [!tip]`\n- Cautions, common mistakes → `> [!warning]`\n- Critical reminders → `> [!important]`\n\n**LaTeX math** (always, never plain ASCII for equations): `$inline$` and `$$display$$`. Use `\\frac`, `\\sin`, `\\cos`, `\\theta`, `\\sqrt`, etc.\n\n**Footnotes** (always for source annotations): `[^1]` markers in text, definitions placed inline near the paragraph that uses them (Obsidian renders them at the bottom regardless, but inline definitions are easier to maintain). Preserve Chinese \u002F non-English annotations verbatim.\n\n**Wikilinks** (for cross-navigation): `[[Note Name]]` between related notes. After the source attribution blockquote, add navigation hints if part of a series:\n```markdown\n> 🔗 **Previous:** [[Chapter N-1]] | **Next:** [[Chapter N+1]]\n```\n\n### 4. Verify before reporting done\n\nAfter writing, do a quick sanity check:\n- Open the file (`Read` it back) and skim — does every section from the source appear?\n- Did you accidentally drop any point list items, exemplar paragraphs, or footnotes?\n- Are there any plain-ASCII equations that should be LaTeX?\n- Did Chinese \u002F non-English annotations get preserved verbatim?\n- Are the callout types appropriate (point list → `[!note]`, not `[!info]`)?\n\nIf the source had 30 numbered points and your output has 28, that's a bug — find the missing two before reporting done.\n\n## Where to save the output\n\nIf the user gave you a specific path, use it. Otherwise, save it next to the source PDF — same folder, same base name with `.md`:\n- `vault\u002Fpath\u002Fchapter1.pdf` → `vault\u002Fpath\u002Fchapter1.md`\n\nIf the existing folder has a `MOC.md` or `00_MOC.md`, also update it to link the new note. See [references\u002Fstyling-rules.md#mocs-and-navigation](references\u002Fstyling-rules.md) for MOC conventions.\n\n## Bulk conversion\n\nIf the user hands you many files at once (e.g., 8 chapter PDFs, all the formula sheets, an entire vault scan), use parallel sub-agents to spread the work. Each agent gets one source file and writes one destination markdown file independently. Brief each agent with:\n- The exact source path (or pre-extracted `.txt` path)\n- The exact destination path\n- A pointer to this skill (`~\u002F.claude\u002Fskills\u002Fobsidian-convert\u002FSKILL.md`) so they apply the same styling\n- Any series-specific wikilinks they should include\n\nWait for all agents to finish, then update the MOC and report.\n\n## Anti-patterns to avoid\n\n- **Summarizing exemplar answers** to \"save space\" — never. These are the model essays the student studies from.\n- **Renumbering point lists** so they don't match the source — preserve original numbering even if it seems weird.\n- **Dropping Chinese annotations** because they look like noise — they're the student's gloss on technical terms.\n- **Skipping image-referenced sections** because the image won't render in markdown — keep the surrounding text and add a `> [!info]` note pointing at the source PDF page.\n- **Writing plain ASCII for equations** (`v = lambda * f`) instead of LaTeX (`$v = \\lambda f$`).\n- **Adding new frontmatter properties** the user's vault doesn't already use — match the conventions of neighbouring notes (run `Read` on a sibling note to check).\n- **Inventing wikilinks to notes that don't exist** — only link `[[Note Name]]` if you've verified the note exists, or if it's clearly a known sibling in a series (Chapter 3 ↔ Chapter 2\u002F4).\n\n## Reference files\n\n- [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md) — full Obsidian Flavored Markdown styling spec for this skill (frontmatter, callouts, LaTeX, footnotes, wikilinks, tables, MOC conventions)\n- [references\u002Fextraction-patterns.md](references\u002Fextraction-patterns.md) — how to extract from text PDFs, image-only PDFs, DOCX, and common gotchas\n","# Obsidian Convert\n\nConvert source files into Obsidian-flavored markdown notes that preserve **all** original content verbatim while applying the user's Obsidian styling conventions.\n\n## Core principle: verbatim fidelity\n\nThe most important rule is that **no content gets lost or paraphrased**. Every numbered point, every exemplar answer, every footnote annotation, every formula, every quoted passage must appear in the output exactly as written in the source. The user is studying from these notes — if you summarize, you've made the note useless for revision.\n\nYou're allowed to **restructure** (split a wall of text into bullets, lift Q&A into callouts, turn comparisons into tables) but you're **not** allowed to **abbreviate**.\n\nThe only things you should strip are pure layout artifacts:\n- Page numbers (e.g., the standalone `1`, `2`, `3` at the bottom of each page)\n- Repeated page headers \u002F running headers (e.g., `吉屋教研组` reappearing at the top of every page)\n- Decorative whitespace between page breaks\n- PDF font-substitution artifacts (orphan ligature glyphs like `fi`, `fl` floating on their own lines — rejoin them into their words)\n\nIf you're ever unsure whether to keep something, keep it.\n\n## Workflow\n\n### 1. Identify the source type and extract text\n\nSource PDFs come in three flavors. Detect which one you have before deciding how to extract.\n\n**Text-extractable PDFs** (the common case — textbook chapters, exam papers, formula sheets typeset in TeX\u002FWord):\n```bash\nmkdir -p \u002Ftmp\u002Fpdf_extracts\npdftotext -layout \"\u003Csource.pdf>\" \u002Ftmp\u002Fpdf_extracts\u002F\u003Cname>.txt\nwc -l \u002Ftmp\u002Fpdf_extracts\u002F\u003Cname>.txt\n```\nIf the extract has more than ~50 lines and you can read text in it (not just garbage), it worked. Read the `.txt` instead of the PDF — bypasses the 32 MB API attachment limit and is much faster.\n\n**Image-only PDFs** (scans, photos, handwritten notes, slide decks with everything rendered as images): `pdftotext` returns \u003C 2 KB of mostly nothing. Fall back to `Read` with the `pages` parameter (max 20 per call):\n```\nRead(\"\u003Csource.pdf>\", pages=\"1-20\")\n```\nRead in chunks if the document is longer.\n\n**DOCX**: try `pandoc -t markdown \u003Cfile.docx>` if pandoc is installed, otherwise note that you couldn't extract and ask the user how to proceed.\n\n### 2. Plan the structure before writing\n\nBefore opening `Write`, skim the entire extracted text and decide:\n- What's the H1 title? (Usually the document's actual title.)\n- What are the major H2 sections? (Chapter sections, topic groupings, syllabus units.)\n- Where are the point lists, exemplar answers, Q&A blocks, tables, equations? Each needs different styling — see [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md).\n- What footnote markers appear? Map them to `[^1]`, `[^2]`… and decide where the definitions live (inline after the paragraph that references them, or all at the bottom).\n- Are there obvious related notes already in the vault to wikilink to? (e.g., a Chapter 3 note should link `[[Chapter 2]]` and `[[Chapter 4]]` for navigation.)\n\nQuick mental check: if you can't answer \"what's the H1 and what are the H2s\" after skimming, you don't understand the doc well enough yet.\n\n### 3. Write the note in one pass\n\nUse the `Write` tool with the full structured markdown. The full styling spec is in [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md) — read it before writing if you're unsure about a callout type or frontmatter convention. The most important conventions:\n\n**Frontmatter** (always):\n```yaml\n---\ndraft: false\ntags:\n  - \u003Csubject>      # e.g., history, physics, economics, chinese\n  - igcse          # or a-level, etc.\n  - \u003Ctopic-slug>   # e.g., chapter1, versailles\naliases:\n  - \u003CAlt name 1>\n  - \u003CChinese name if applicable>\n---\n```\n\n**Title line** (always): `# \u003Ctitle from source>`\n\n**Source attribution blockquote** (always, right after the H1):\n```markdown\n> Source PDF: `\u003Cfilename.pdf>`. \u003Cone-line description of what's in the doc>.\n```\n\n**Callout mapping** (see styling-rules.md for the full table):\n- Point lists → `> [!note]- Point List` (collapsed)\n- Exemplar \u002F model answers → `> [!example]- Exemplar Answer` (collapsed)\n- Definitions → `> [!def]`\n- Quoted source text \u002F poems \u002F passages to memorize → `> [!quote]`\n- Q&A discussion questions → `> [!question]`\n- Sidebar info, related-notes lists → `> [!info]`\n- Strategy tips, exam advice → `> [!tip]`\n- Cautions, common mistakes → `> [!warning]`\n- Critical reminders → `> [!important]`\n\n**LaTeX math** (always, never plain ASCII for equations): `$inline$` and `$$display$$`. Use `\\frac`, `\\sin`, `\\cos`, `\\theta`, `\\sqrt`, etc.\n\n**Footnotes** (always for source annotations): `[^1]` markers in text, definitions placed inline near the paragraph that uses them (Obsidian renders them at the bottom regardless, but inline definitions are easier to maintain). Preserve Chinese \u002F non-English annotations verbatim.\n\n**Wikilinks** (for cross-navigation): `[[Note Name]]` between related notes. After the source attribution blockquote, add navigation hints if part of a series:\n```markdown\n> 🔗 **Previous:** [[Chapter N-1]] | **Next:** [[Chapter N+1]]\n```\n\n### 4. Verify before reporting done\n\nAfter writing, do a quick sanity check:\n- Open the file (`Read` it back) and skim — does every section from the source appear?\n- Did you accidentally drop any point list items, exemplar paragraphs, or footnotes?\n- Are there any plain-ASCII equations that should be LaTeX?\n- Did Chinese \u002F non-English annotations get preserved verbatim?\n- Are the callout types appropriate (point list → `[!note]`, not `[!info]`)?\n\nIf the source had 30 numbered points and your output has 28, that's a bug — find the missing two before reporting done.\n\n## Where to save the output\n\nIf the user gave you a specific path, use it. Otherwise, save it next to the source PDF — same folder, same base name with `.md`:\n- `vault\u002Fpath\u002Fchapter1.pdf` → `vault\u002Fpath\u002Fchapter1.md`\n\nIf the existing folder has a `MOC.md` or `00_MOC.md`, also update it to link the new note. See [references\u002Fstyling-rules.md#mocs-and-navigation](references\u002Fstyling-rules.md) for MOC conventions.\n\n## Bulk conversion\n\nIf the user hands you many files at once (e.g., 8 chapter PDFs, all the formula sheets, an entire vault scan), use parallel sub-agents to spread the work. Each agent gets one source file and writes one destination markdown file independently. Brief each agent with:\n- The exact source path (or pre-extracted `.txt` path)\n- The exact destination path\n- A pointer to this skill (`~\u002F.claude\u002Fskills\u002Fobsidian-convert\u002FSKILL.md`) so they apply the same styling\n- Any series-specific wikilinks they should include\n\nWait for all agents to finish, then update the MOC and report.\n\n## Anti-patterns to avoid\n\n- **Summarizing exemplar answers** to \"save space\" — never. These are the model essays the student studies from.\n- **Renumbering point lists** so they don't match the source — preserve original numbering even if it seems weird.\n- **Dropping Chinese annotations** because they look like noise — they're the student's gloss on technical terms.\n- **Skipping image-referenced sections** because the image won't render in markdown — keep the surrounding text and add a `> [!info]` note pointing at the source PDF page.\n- **Writing plain ASCII for equations** (`v = lambda * f`) instead of LaTeX (`$v = \\lambda f$`).\n- **Adding new frontmatter properties** the user's vault doesn't already use — match the conventions of neighbouring notes (run `Read` on a sibling note to check).\n- **Inventing wikilinks to notes that don't exist** — only link `[[Note Name]]` if you've verified the note exists, or if it's clearly a known sibling in a series (Chapter 3 ↔ Chapter 2\u002F4).\n\n## Reference files\n\n- [references\u002Fstyling-rules.md](references\u002Fstyling-rules.md) — full Obsidian Flavored Markdown styling spec for this skill (frontmatter, callouts, LaTeX, footnotes, wikilinks, tables, MOC conventions)\n- [references\u002Fextraction-patterns.md](references\u002Fextraction-patterns.md) — how to extract from text PDFs, image-only PDFs, DOCX, and common gotchas\n",{"name":5,"description":7},"main","6ef866230c1ba3459fd08fc65141dfd37fffd41a",0,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills","https:\u002F\u002Fskills.henryzh.dev","https:\u002F\u002Fgithub.com\u002Fcaezium","henryzh.dev","quant, data science, full-stack, robotics.\r\nbuilding cool stuff","2026-08-24 19:21:44",{"source":8,"skillName":5,"commands":59,"prompt":69,"zipAvailable":70},[60,63,66],{"tool":61,"command":62},"npx","npx skills add caezium\u002Fskills --skill obsidian-convert",{"tool":64,"command":65},"bunx","bunx skills add caezium\u002Fskills --skill obsidian-convert",{"tool":67,"command":68},"pnpm","pnpm dlx skills add caezium\u002Fskills --skill obsidian-convert","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fobsidian-convert\u002FSKILL.md",true,[72,78],{"path":73,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":76,"sourceHtmlUrl":77},"references\u002Fextraction-patterns.md","REFERENCE","参考文档",5425,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fobsidian-convert\u002Freferences\u002Fextraction-patterns.md",{"path":79,"resourceType":74,"resourceTypeDesc":75,"sizeBytes":80,"sourceHtmlUrl":81},"references\u002Fstyling-rules.md",12072,"https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fobsidian-convert\u002Freferences\u002Fstyling-rules.md",[83,104,132,150,170,192],{"skillNo":84,"name":85,"nameZh":6,"description":86,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":87,"categories":88,"keywords":97,"publishedAt":44},"01M71JT4RYCAS18RF6003YYR1D","writing-for-agents","Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwriting-for-agents\u002FSKILL.md",[89,92],{"categoryNo":90,"categoryName":91,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":93,"categoryName":94,"taxonomyType":95,"taxonomyTypeDesc":96},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[98,99,100,101,102,103,35],"writing for agents","for","writing","editing","use","documents",{"skillNo":105,"name":106,"nameZh":6,"description":107,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":108,"categories":109,"keywords":113,"publishedAt":44},"01M71JT0GES4AZ5ADH8EADNN1Z","wrangler","Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwrangler\u002FSKILL.md",[110],{"categoryNo":111,"categoryName":112,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000010","后端与接口",[106,114,115,116,117,118,119,120,121,122,123,124,125,99,126,127,128,129,130,131],"cloudflare","d1","knowledge","workflows","cli","secrets","commands","syntax","cloudflare workers","r2","queues","running","containers","correct","load","workers","best practices","store",{"skillNo":133,"name":134,"nameZh":6,"description":135,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":136,"categories":137,"keywords":138,"publishedAt":44},"01M71JSWX4R203M32K9A18AAW8","workers-best-practices","Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fworkers-best-practices\u002FSKILL.md",[],[139,129,130,140,141,142,99,114,143,116,144,106,100,128,35,145,146,147,148,149],"workers best practices","observability","reviews","promises","worker","new","code","common","streaming","state","anti-patterns",{"skillNo":151,"name":152,"nameZh":6,"description":153,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":154,"categories":155,"keywords":156,"publishedAt":44},"01M71JST6P88MXGT2MWDTPJGEG","wizard","Generate an interactive bash wizard that walks a human through steps only they can perform. Use when provisioning infrastructure, setting up credentials or CI secrets, walking an unfamiliar third-party dashboard, or running a one-off migration or cutover. Don't invoke this for steps the agent can perform itself.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwizard\u002FSKILL.md",[],[152,125,157,99,158,159,160,38,102,161,35,162,163,164,165,166,167,168,169,119],"setting","up","ci","provisioning infrastructure","invoke","Bash","one","interactive","can","credentials","dashboard","generate","human",{"skillNo":171,"name":172,"nameZh":6,"description":173,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":174,"sourceHtmlUrl":175,"categories":176,"keywords":177,"publishedAt":44},"01M71JSQ95S7BQ61DAPN95X2W4","why","Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available evidence sources, searches each relevant category, and returns a cited read on decisions and tradeoffs. Use how for runtime behavior.","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fwhy\u002FSKILL.md",[],[172,178,179,180,181,99,182,183,102,184,185,186,187,188,189,190,191],"sources","thresholds","evidence","relevant","runtime","design","decisions","work","how","behavior","read","data","rationale","returns",{"skillNo":193,"name":194,"nameZh":6,"description":195,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":196,"categories":197,"keywords":201,"publishedAt":44},"01M71JSMHFVSH8HK8GXN6YAX6J","week-prep","Generate a briefing with calendar, tasks, and attention flags. Use for \"week prep\", \"briefing\", or \"what's on this week\".","https:\u002F\u002Fgithub.com\u002Fcaezium\u002Fskills\u002Fblob\u002Fmain\u002Fweek-prep\u002FSKILL.md",[198],{"categoryNo":199,"categoryName":200,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发",[202,203,204,205,206,99,207,208,209,210,102,162,211,212,213,168,214,188,215,216],"week prep","prep","deep","attention","get","search","messages","briefing","MCP","tasks","with","slack","calendar","flags","grep"]