[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AYDT43DEMYC3XZ9JHS894K":3,"skill-same:01M7AYDT43DEMYC3XZ9JHS894K":76},{"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":41,"commitSha":42,"skillDirectory":43,"forksCount":44,"repositoryHtmlUrl":45,"repositoryHomepageUrl":6,"ownerHtmlUrl":46,"ownerBlogUrl":47,"ownerDescription":48,"updatedAt":49,"install":50,"resources":63},"01M7AYDT43DEMYC3XZ9JHS894K","narrative-book-summaries",null,"Use when summarizing narrative\u002Fargument books into vaults.","oat431\u002Fhermes_config_backup","oat431","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F54722120?v=4","01M7AYD07EH8SYPMYY8CGZKS22","Oralita","01M7AYD0816QDF77E3TWCRSQC9",1,"MIT","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fbook-summarizer\u002Fnarrative-book-summaries\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: narrative-book-summaries\ndescription: \"Use when summarizing narrative\u002Fargument books into vaults.\"\nversion: 1.0.0\nauthor: curator\nlicense: MIT\nplatforms: [windows, linux, macos]\nmetadata:\n  hermes:\n    tags: [books, summarization, obsidian, mermaid, markdown-style]\n    related_skills: [oralita-book-sum-obs, obsidian, research-paper-summaries]\n---\n\n# Narrative Book → Obsidian Study Vault\n\n## When to Use\n\nArgument-driven books where each chapter is a self-contained essay with a thesis: big history, popular science, philosophy, case-study narratives. For technical\u002Fpractice books, use `oralita-book-sum-obs` (the execution engine: extraction, manifest, delegation, verification) with its study templates. That skill is user-protected; this companion skill carries the narrative-book FORMATS the user has approved and the style rules that apply to everything generated. For research papers (single-file study notes with quote and number verification), use `research-paper-summaries`.\n\nSignals: 10–20+ titled chapters with distinct arguments; the user asks to \"plan the format first\" because the book is \"different from other books\"; a prior narrative-book vault exists to mirror (e.g. `psychology_book\u002Fschema_therapy`).\n\n## Approved Note Anatomy (user-approved 2026-09-03, Sapiens run)\n\nOne file per chapter, `NN_Title_With_Underscores.md`, plus `Book Title - Overview.md`:\n\n- YAML frontmatter: `tags: [booktag, part-tag, 1-3 topic tags, history]` — no `title:` field.\n- Source blockquote: `> *Source: \u003CTitle> by \u003CAuthor> (\u003Cpublisher, year>), Chapter N \"\u003Cexact title>\", PDF pp. A–B (printed pp. unknown)*`\n- H1: `# NN: \u003CExact Chapter Title>` (colon, never em-dash)\n- `## Key Idea` — the chapter's thesis + why it matters, 2–3 paragraphs (~250–350 words), mentor voice.\n- `## The Argument` — bold-led bullets reconstructing the evidence spine in own words.\n- `## Key Concepts` — `### Name` per durable concept: define + example + why it matters. The book's terminology.\n- `## Memorable Quotes` — 2–4 short verbatim quotes, each `> \"...\" (PDF p. N)`.\n- `## Connections` — `- [[Stem]]: one-line reason` wikilinks to sibling notes + the overview last. Colon separator, never em-dash.\n- `## Takeaway` — one line worth remembering.\n\nFull spec, per-section word budgets, and the Sapiens worked example: `references\u002Fnarrative-note-anatomy.md`.\n\n## Style Rules (embed from the start — user preferences)\n\n- **No em-dashes in generated prose.** Use commas, colons, parentheses, or split sentences. En-dashes stay in page ranges (`PDF pp. 84–104`).\n- **Verbatim quotes are sacred.** Never alter quoted punctuation, including short quoted fragments inside prose (skip them in any bulk punctuation edit).\n- Mermaid node labels with special chars must be quoted (`A[\"text (parens)\"]`); edge labels with special chars quoted as `-->|\"label ()\u002F\"|`.\n- Match the destination vault's naming\u002Flink conventions over any global rule — underscore stems are fine where the local reference vault uses them.\n- Overview\u002FMOC: chapter-map table grouped by part with PDF-page sources, one Mermaid argument map, reading paths (goal → start notes), 5–9 core ideas, honest footer with extraction limitations.\n\n## Workflow\n\n1. **Plan first when asked.** Inspect the PDF (pymupdf `get_toc()`, page count, text-layer vs OCR check), read 1–3 reference-vault notes, then present the concrete plan: per-chapter anatomy, citation scheme, file list. Confirm granularity and quote sections with the user before writing anything.\n2. **Extract with page markers.** Insert a `[PDF p. N]` marker at every PDF-page boundary inside each chapter extract. Writers then attribute quotes from markers, and validation can check attribution mechanically. Verify chapter boundaries by reading head\u002Ftail text of each extract, not the TOC alone.\n3. **Delegate in live-limited batches.** Query `delegation.max_concurrent_children` at run time. Every writer gets: extract path, staging path, format spec, an exemplar note, and the complete canonical link allow-list. A child's report is NOT proof — validate every staging file (frontmatter, source line, section order, quotes verbatim + page attribution, links) before promotion. Use `scripts\u002Fcheck_chapter_notes.py`.\n4. **Promote, verify, spot-check.** Promote validated files, run the vault verifier, then spot-check 5+ factual claims per representative note against the source text (grep with normalized whitespace — extracts are line-wrapped, so phrase search must flatten whitespace first; PDF text layers also hyphenate across lines (\"develop-\\nment\"), so also try the \"- \"-joined form before concluding a quote is absent).\n5. **Compression pass (optional).** The user may ask for shorter notes after the first draft. Rules in `references\u002Fnarrative-note-anatomy.md`.\n\n## Mermaid Diagram Rubric\n\nAdd a diagram only when the chapter's structure IS a relationship:\n\n- **Strong candidates:** cycles\u002Ffeedback loops (luxury trap, capitalist circle), abstraction ladders (money), taxonomies (religion tree), branching futures (three paths), vicious circles as `stateDiagram-v2`.\n- **Good candidates:** ontologies (objective\u002Fsubjective\u002Finter-subjective), imperial cycles, self-defeating prophecy (`stateDiagram-v2`), science↔politics↔economics loops, three-way alliances.\n- **Skip:** plain lists, debates without a spine, psychological findings. Never force a sequence diagram; use `stateDiagram-v2` where the structure is genuinely stateful.\n- Place each diagram inside the Key Concept it illustrates.\n- After editing, verify: code fences balanced, square brackets matched, vault verifier passes.\n\n## Pitfalls\n\n- **Fuzzy patch can over-match.** A long `old_string` that partially matches may swallow a preceding sentence (observed: a patch deleted the opening sentence of a concept paragraph). Check every diff and restore immediately; prefer short unique anchors (heading lines) over long paragraph anchors.\n- **patch escape-drift.** Do not backslash-escape double quotes in patch strings; plain quotes match plain quotes in the file. If a match fails with \"Did you mean\", re-grep the exact glyphs (apostrophes, dashes) before retrying.\n- **Calibre\u002FPDFDrive conversions have no folio numbers.** Scan for digit-only lines to detect printed page numbers; if absent, cite PDF pages and mark `printed pp. unknown` everywhere. Never invent printed ranges.\n- **Endnote texts are usually absent from e-book extracts** — never assert footnote content.\n- **Homo Deus (Sapiens sequel) is queued** in `F:\u002Fbooks\u002Fhuman_book\u002F` — reuse this anatomy and the Sapiens vault as the format reference.\n\n## Reference Map\n\n- `references\u002Fnarrative-note-anatomy.md` — full per-section spec, compression-pass rules, Sapiens case study incl. the 14-diagram map.\n- `scripts\u002Fcheck_chapter_notes.py` — validates staging\u002Fvault notes: frontmatter, source line, H1, section order, verbatim quotes + page attribution, wikilink allow-list.","# Narrative Book → Obsidian Study Vault\n\n## When to Use\n\nArgument-driven books where each chapter is a self-contained essay with a thesis: big history, popular science, philosophy, case-study narratives. For technical\u002Fpractice books, use `oralita-book-sum-obs` (the execution engine: extraction, manifest, delegation, verification) with its study templates. That skill is user-protected; this companion skill carries the narrative-book FORMATS the user has approved and the style rules that apply to everything generated. For research papers (single-file study notes with quote and number verification), use `research-paper-summaries`.\n\nSignals: 10–20+ titled chapters with distinct arguments; the user asks to \"plan the format first\" because the book is \"different from other books\"; a prior narrative-book vault exists to mirror (e.g. `psychology_book\u002Fschema_therapy`).\n\n## Approved Note Anatomy (user-approved 2026-09-03, Sapiens run)\n\nOne file per chapter, `NN_Title_With_Underscores.md`, plus `Book Title - Overview.md`:\n\n- YAML frontmatter: `tags: [booktag, part-tag, 1-3 topic tags, history]` — no `title:` field.\n- Source blockquote: `> *Source: \u003CTitle> by \u003CAuthor> (\u003Cpublisher, year>), Chapter N \"\u003Cexact title>\", PDF pp. A–B (printed pp. unknown)*`\n- H1: `# NN: \u003CExact Chapter Title>` (colon, never em-dash)\n- `## Key Idea` — the chapter's thesis + why it matters, 2–3 paragraphs (~250–350 words), mentor voice.\n- `## The Argument` — bold-led bullets reconstructing the evidence spine in own words.\n- `## Key Concepts` — `### Name` per durable concept: define + example + why it matters. The book's terminology.\n- `## Memorable Quotes` — 2–4 short verbatim quotes, each `> \"...\" (PDF p. N)`.\n- `## Connections` — `- [[Stem]]: one-line reason` wikilinks to sibling notes + the overview last. Colon separator, never em-dash.\n- `## Takeaway` — one line worth remembering.\n\nFull spec, per-section word budgets, and the Sapiens worked example: `references\u002Fnarrative-note-anatomy.md`.\n\n## Style Rules (embed from the start — user preferences)\n\n- **No em-dashes in generated prose.** Use commas, colons, parentheses, or split sentences. En-dashes stay in page ranges (`PDF pp. 84–104`).\n- **Verbatim quotes are sacred.** Never alter quoted punctuation, including short quoted fragments inside prose (skip them in any bulk punctuation edit).\n- Mermaid node labels with special chars must be quoted (`A[\"text (parens)\"]`); edge labels with special chars quoted as `-->|\"label ()\u002F\"|`.\n- Match the destination vault's naming\u002Flink conventions over any global rule — underscore stems are fine where the local reference vault uses them.\n- Overview\u002FMOC: chapter-map table grouped by part with PDF-page sources, one Mermaid argument map, reading paths (goal → start notes), 5–9 core ideas, honest footer with extraction limitations.\n\n## Workflow\n\n1. **Plan first when asked.** Inspect the PDF (pymupdf `get_toc()`, page count, text-layer vs OCR check), read 1–3 reference-vault notes, then present the concrete plan: per-chapter anatomy, citation scheme, file list. Confirm granularity and quote sections with the user before writing anything.\n2. **Extract with page markers.** Insert a `[PDF p. N]` marker at every PDF-page boundary inside each chapter extract. Writers then attribute quotes from markers, and validation can check attribution mechanically. Verify chapter boundaries by reading head\u002Ftail text of each extract, not the TOC alone.\n3. **Delegate in live-limited batches.** Query `delegation.max_concurrent_children` at run time. Every writer gets: extract path, staging path, format spec, an exemplar note, and the complete canonical link allow-list. A child's report is NOT proof — validate every staging file (frontmatter, source line, section order, quotes verbatim + page attribution, links) before promotion. Use `scripts\u002Fcheck_chapter_notes.py`.\n4. **Promote, verify, spot-check.** Promote validated files, run the vault verifier, then spot-check 5+ factual claims per representative note against the source text (grep with normalized whitespace — extracts are line-wrapped, so phrase search must flatten whitespace first; PDF text layers also hyphenate across lines (\"develop-\\nment\"), so also try the \"- \"-joined form before concluding a quote is absent).\n5. **Compression pass (optional).** The user may ask for shorter notes after the first draft. Rules in `references\u002Fnarrative-note-anatomy.md`.\n\n## Mermaid Diagram Rubric\n\nAdd a diagram only when the chapter's structure IS a relationship:\n\n- **Strong candidates:** cycles\u002Ffeedback loops (luxury trap, capitalist circle), abstraction ladders (money), taxonomies (religion tree), branching futures (three paths), vicious circles as `stateDiagram-v2`.\n- **Good candidates:** ontologies (objective\u002Fsubjective\u002Finter-subjective), imperial cycles, self-defeating prophecy (`stateDiagram-v2`), science↔politics↔economics loops, three-way alliances.\n- **Skip:** plain lists, debates without a spine, psychological findings. Never force a sequence diagram; use `stateDiagram-v2` where the structure is genuinely stateful.\n- Place each diagram inside the Key Concept it illustrates.\n- After editing, verify: code fences balanced, square brackets matched, vault verifier passes.\n\n## Pitfalls\n\n- **Fuzzy patch can over-match.** A long `old_string` that partially matches may swallow a preceding sentence (observed: a patch deleted the opening sentence of a concept paragraph). Check every diff and restore immediately; prefer short unique anchors (heading lines) over long paragraph anchors.\n- **patch escape-drift.** Do not backslash-escape double quotes in patch strings; plain quotes match plain quotes in the file. If a match fails with \"Did you mean\", re-grep the exact glyphs (apostrophes, dashes) before retrying.\n- **Calibre\u002FPDFDrive conversions have no folio numbers.** Scan for digit-only lines to detect printed page numbers; if absent, cite PDF pages and mark `printed pp. unknown` everywhere. Never invent printed ranges.\n- **Endnote texts are usually absent from e-book extracts** — never assert footnote content.\n- **Homo Deus (Sapiens sequel) is queued** in `F:\u002Fbooks\u002Fhuman_book\u002F` — reuse this anatomy and the Sapiens vault as the format reference.\n\n## Reference Map\n\n- `references\u002Fnarrative-note-anatomy.md` — full per-section spec, compression-pass rules, Sapiens case study incl. the 14-diagram map.\n- `scripts\u002Fcheck_chapter_notes.py` — validates staging\u002Fvault notes: frontmatter, source line, H1, section order, verbatim quotes + page attribution, wikilink allow-list.",{"name":5,"author":24,"license":15,"version":25,"metadata":26,"platforms":37,"description":7},"curator","1.0.0",{"hermes":27},{"tags":28,"related_skills":34},[29,30,31,32,33],"books","summarization","obsidian","mermaid","markdown-style",[35,31,36],"oralita-book-sum-obs","research-paper-summaries",[38,39,40],"windows","linux","macos","main","2988869c5a7d7be2225c9ea7a54a4a56a59b0d45","workflow\u002Fbook-summarizer\u002Fnarrative-book-summaries",0,"https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup","https:\u002F\u002Fgithub.com\u002Foat431","https:\u002F\u002Fpanomete.com\u002F","Hope and Luck is not a part of the strategy, \r\n\r\nSoftskill produces plan Hard skill produce product.","2026-11-03 00:43:08",{"source":8,"skillName":5,"commands":51,"prompt":61,"zipAvailable":62},[52,55,58],{"tool":53,"command":54},"npx","npx skills add oat431\u002Fhermes_config_backup --skill narrative-book-summaries",{"tool":56,"command":57},"bunx","bunx skills add oat431\u002Fhermes_config_backup --skill narrative-book-summaries",{"tool":59,"command":60},"pnpm","pnpm dlx skills add oat431\u002Fhermes_config_backup --skill narrative-book-summaries","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fbook-summarizer\u002Fnarrative-book-summaries\u002FSKILL.md",true,[64,70],{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"references\u002Fnarrative-note-anatomy.md","REFERENCE","参考文档",4152,"https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fbook-summarizer\u002Fnarrative-book-summaries\u002Freferences\u002Fnarrative-note-anatomy.md",{"path":71,"resourceType":72,"resourceTypeDesc":73,"sizeBytes":74,"sourceHtmlUrl":75},"scripts\u002Fcheck_chapter_notes.py","SCRIPT","脚本",4699,"https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fbook-summarizer\u002Fnarrative-book-summaries\u002Fscripts\u002Fcheck_chapter_notes.py",[77,84,91,99,106,113],{"skillNo":78,"name":79,"nameZh":6,"description":80,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":81,"categories":82,"keywords":83,"publishedAt":19},"01M7AYZ2JY3NFKAPQSC39N5F3F","project-spec-docs","Fill\u002Fupdate project_spec docs (templates, MM handoffs).","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fproject-spec-docs\u002FSKILL.md",[],[],{"skillNo":85,"name":86,"nameZh":6,"description":87,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":88,"categories":89,"keywords":90,"publishedAt":19},"01M7AYYRZPE7HQX7778558X52S","penpot-mcp","Penpot MCP: connect, verify, and fix streaming via nginx.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fpenpot-mcp\u002FSKILL.md",[],[],{"skillNo":92,"name":93,"nameZh":6,"description":94,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":95,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7AYYNKP61CN2281VMJDXMB5","nginx-streaming-proxy","Configure nginx for streaming endpoints (SSE, MCP, WS).","Unlicense","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fnginx-streaming-proxy\u002FSKILL.md",[],[],{"skillNo":100,"name":101,"nameZh":6,"description":102,"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},"01M7AYYJP0MDV1D3WVEEZB3SGA","branding-ci","Use for brand\u002FCI identity, style guides, and brand docs.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fui-ux\u002Fbranding-ci\u002FSKILL.md",[],[],{"skillNo":107,"name":108,"nameZh":6,"description":109,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":95,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":19},"01M7AYYGCXVT97DDTCM0ZJT6F6","yuanbao","Yuanbao (元宝) groups: @mention users, query info\u002Fmembers.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fshared-skills\u002Fyuanbao\u002FSKILL.md",[],[],{"skillNo":114,"name":115,"nameZh":6,"description":116,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":95,"licenseFileUrl":6,"sourceHtmlUrl":117,"categories":118,"keywords":119,"publishedAt":19},"01M7AYYC2Y65NMGMD4GQXAZMMS","slidev","Create and present web-based slidedecks for developers using Slidev with Markdown, Vue components, code highlighting, animations, and interactive features. Use when building technical presentations, conference talks, code walkthroughs, teaching materials, or developer decks.","https:\u002F\u002Fgithub.com\u002Foat431\u002Fhermes_config_backup\u002Fblob\u002Fmain\u002Fworkflow\u002Fshared-skills\u002Fslidev\u002FSKILL.md",[],[]]