book-analyzer

Name: book-analyzer
Author: tuan3w

32stars2forks

How to install book-analyzer?

tuan3w/obsidian-vault-agent/book-analyzer

$npx -y skills add tuan3w/obsidian-vault-agent --skill book-analyzer

Installs into the current project.

Summary

Analyze a book (EPUB/PDF) and generate detailed chapter-by-chapter notes with extracted key concepts. Use when given a book file path to process.

Files · 1

View on GitHub

SKILL.md

1	<Purpose>
2	Full autonomous analysis of book files (EPUB/PDF) into structured notes.
3	Extracts text, detects chapters, analyzes each chapter via parallel agents,
4	synthesizes across chapters, and optionally integrates into an Obsidian vault
5	with wikilinks and extracted Term notes.
6	</Purpose>
7
8	<Use_When>
9	- User provides a path to an EPUB or PDF book file
10	- User says "analyze this book", "process this book", "read this book"
11	- User wants structured notes from a book-length document (20+ pages)
12	</Use_When>
13
14	<Do_Not_Use_When>
15	- Short documents under 20 pages — use /paper or /process instead
16	- User wants to read or search specific sections — just use Read tool directly
17	- User wants EPUB output from markdown — wrong direction
18	- URL-only input with no file on disk — use /research instead
19	</Do_Not_Use_When>
20
21	<Why_This_Exists>
22	Book notes that merely transcribe content are useless — the value is in synthesis.
23	Most book notes in a vault end up sparse because processing a full book manually is
24	exhausting. This skill automates the mechanical work (extraction, splitting, formatting)
25	while delegating the intellectual work (synthesis, assessment, connections) to specialized
26	agents with strict quality constraints.
27	</Why_This_Exists>
28
29	<Execution_Policy>
30	- Run extraction scripts first, verify output before proceeding to analysis
31	- Delegate chapter analysis to parallel agents (one per chapter, all simultaneous)
32	- Use Sonnet for chapter analysis, Opus for cross-book synthesis
33	- Vault integration is conditional — detect .obsidian/ or CLAUDE.md in working directory
34	- Report progress via TodoWrite at each stage transition
35	- If extraction fails, stop and report — don't guess at content
36	- Total chapters capped at 50 — if more, group adjacent chapters into batches
37	</Execution_Policy>
38
39	<Steps>
40
41	## Stage 1: PARSE ARGUMENTS AND EXTRACT
42
43	Parse `$ARGUMENTS` to get the book file path and optional flags.
44
45	Argument format:
46	```
47	$ARGUMENTS = "path/to/book.epub" # basic
48	$ARGUMENTS = "path/to/book.pdf --no-terms" # skip term extraction
49	$ARGUMENTS = "path/to/book.epub --output ~/Desktop/" # custom output location
50	```
51
52	Parse logic:
53	1. Split $ARGUMENTS by spaces, treating quoted paths as single tokens
54	2. First non-flag token = file path (REQUIRED — stop if missing)
55	3. Optional flags: `--no-terms` (skip concept extraction), `--output <dir>` (custom output)
56	4. Detect format from file extension: `.epub`, `.pdf`, `.mobi`
57
58	Run extraction based on format:
59
60	For EPUB:
61	```bash
62	SKILL_DIR="$(dirname "$(readlink -f "$0" 2>/dev/null \|\| echo "$0")")"
63	# Find skill directory — check common locations
64	for dir in .claude/skills/book-analyzer ~/.claude/skills/book-analyzer; do
65	if [ -f "$dir/scripts/extract_epub.py" ]; then SKILL_DIR="$dir"; break; fi
66	done
67	WORK_DIR=".book-work-$(date +%s)"
68	python3 "$SKILL_DIR/scripts/extract_epub.py" "INPUT_PATH" "$WORK_DIR"
69	```
70
71	For PDF:
72	```bash
73	WORK_DIR=".book-work-$(date +%s)"
74	"$SKILL_DIR/scripts/extract_pdf.sh" "INPUT_PATH" "$WORK_DIR"
75	```
76
77	For MOBI: Convert to EPUB first with `ebook-convert` (calibre), then run EPUB extraction.
78
79	The work directory is created inside the vault root (e.g., `.book-work-1709398200/`).
80	Clean it up after the pipeline completes.
81
82	After extraction: Read `$WORK_DIR/metadata.json` to verify success.
83
84	PDF fallback: If metadata shows `"needs_fallback": true`, use the Claude Read tool instead:
85	- Read the PDF 20 pages at a time: `Read(file_path="INPUT_PATH", pages="1-20")`, then `pages="21-40"`, etc.
86	- Write each batch to `$WORK_DIR/pages/page_batch_NNN.txt`
87	- Update metadata with actual content
88
89	## Stage 2: STRUCTURE
90
91	Read `metadata.json` from the extraction output.
92
93	For EPUBs: Chapter structure comes directly from extraction (H1/H2 splitting). Proceed to Stage 3.
94
95	For PDFs: Chapter detection depends on extraction quality.
96	- If pages have clear chapter headings, group pages into chapters
97	- If no clear structure: Launch a single agent to read the first 5 pages and detect chapter boundaries:
98
99	```
100	Agent(
101	subagent_type="general-purpose",
102	model="haiku",
103	prompt="Read the following text from the first 5 pages of a book. Identify chapter
104	boundaries and return a JSON array of {title, start_page, end_page} objects.
105	If no chapters are detectable, return [{title: 'Full Text', start_page: 1, end_page: LAST}].
106
107	TEXT:
108	[first 5 pages content]"
109	)
110	```
111
112	- Update metadata with detected chapter list
113	- Group page files into chapter files by concatenation
114
115	Result: A `chapters/` directory with one file per chapter, and updated `metadata.json`.
116
117	## Stage 3: ANALYZE (parallel, file-based)
118
119	Read the agent definition from `agents/

People who liked this also viewed

6 picks

Type

grill-with-docsA relentless interview to sharpen a plan or design, which also creates docs (ADR's and glossary) as we go.SkillsJul 2026537k182k
lark-minutes飞书妙记：搜索妙记、查看妙记基础信息、下载/上传音视频、读取或编辑妙记的产物内容、改标题、替换说话人/关键词、申请妙记查看/编辑权限。当给出minute_token、本地音视频文件，要查/改/转妙记产物，或用户明确要主动申请妙记权限时使用；本地音视频转纪要/逐字稿优先走本 skill，不要用 ffmpeg/whisper 本地转写。不负责：获取会议关联妙记，或仅按自然语言标题定位纪要SkillsJul 2026379k16k
lark-markdown飞书 Markdown：查看、创建、上传、编辑和比较 Markdown 文件。当用户需要创建或编辑 Markdown 文件、读取、修改、局部 patch 或比较差异时使用。不负责将 Markdown 导入为飞书在线文档，也不负责文件搜索、权限、评论、移动、删除等云空间管理操作。SkillsJul 2026295k16k
docxUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', SkillsJul 2026157k163k
edit-articleSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026154k181k
lark-note飞书会议纪要（Note）直查：已知 note_id 时查询纪要详情、展示类型、关联文档 token，并读取 unified 原始逐字记录。当用户已持有 note_id，或从文档显式 vc-node-id 获得 note_id 时使用。不负责会议/日程/妙记定位、文档标题搜索或 Docx 正文读取。SkillsJul 2026154k16k

book-analyzer

32stars2forks

View on GitHub

How to install book-analyzer?

tuan3w/obsidian-vault-agent/book-analyzer

$npx -y skills add tuan3w/obsidian-vault-agent --skill book-analyzer

Installs into the current project.

Summary

Analyze a book (EPUB/PDF) and generate detailed chapter-by-chapter notes with extracted key concepts. Use when given a book file path to process.

Files · 1

View on GitHub

SKILL.md

1	<Purpose>
2	Full autonomous analysis of book files (EPUB/PDF) into structured notes.
3	Extracts text, detects chapters, analyzes each chapter via parallel agents,
4	synthesizes across chapters, and optionally integrates into an Obsidian vault
5	with wikilinks and extracted Term notes.
6	</Purpose>
7
8	<Use_When>
9	- User provides a path to an EPUB or PDF book file
10	- User says "analyze this book", "process this book", "read this book"
11	- User wants structured notes from a book-length document (20+ pages)
12	</Use_When>
13
14	<Do_Not_Use_When>
15	- Short documents under 20 pages — use /paper or /process instead
16	- User wants to read or search specific sections — just use Read tool directly
17	- User wants EPUB output from markdown — wrong direction
18	- URL-only input with no file on disk — use /research instead
19	</Do_Not_Use_When>
20
21	<Why_This_Exists>
22	Book notes that merely transcribe content are useless — the value is in synthesis.
23	Most book notes in a vault end up sparse because processing a full book manually is
24	exhausting. This skill automates the mechanical work (extraction, splitting, formatting)
25	while delegating the intellectual work (synthesis, assessment, connections) to specialized
26	agents with strict quality constraints.
27	</Why_This_Exists>
28
29	<Execution_Policy>
30	- Run extraction scripts first, verify output before proceeding to analysis
31	- Delegate chapter analysis to parallel agents (one per chapter, all simultaneous)
32	- Use Sonnet for chapter analysis, Opus for cross-book synthesis
33	- Vault integration is conditional — detect .obsidian/ or CLAUDE.md in working directory
34	- Report progress via TodoWrite at each stage transition
35	- If extraction fails, stop and report — don't guess at content
36	- Total chapters capped at 50 — if more, group adjacent chapters into batches
37	</Execution_Policy>
38
39	<Steps>
40
41	## Stage 1: PARSE ARGUMENTS AND EXTRACT
42
43	Parse `$ARGUMENTS` to get the book file path and optional flags.
44
45	Argument format:
46	```
47	$ARGUMENTS = "path/to/book.epub" # basic
48	$ARGUMENTS = "path/to/book.pdf --no-terms" # skip term extraction
49	$ARGUMENTS = "path/to/book.epub --output ~/Desktop/" # custom output location
50	```
51
52	Parse logic:
53	1. Split $ARGUMENTS by spaces, treating quoted paths as single tokens
54	2. First non-flag token = file path (REQUIRED — stop if missing)
55	3. Optional flags: `--no-terms` (skip concept extraction), `--output <dir>` (custom output)
56	4. Detect format from file extension: `.epub`, `.pdf`, `.mobi`
57
58	Run extraction based on format:
59
60	For EPUB:
61	```bash
62	SKILL_DIR="$(dirname "$(readlink -f "$0" 2>/dev/null \|\| echo "$0")")"
63	# Find skill directory — check common locations
64	for dir in .claude/skills/book-analyzer ~/.claude/skills/book-analyzer; do
65	if [ -f "$dir/scripts/extract_epub.py" ]; then SKILL_DIR="$dir"; break; fi
66	done
67	WORK_DIR=".book-work-$(date +%s)"
68	python3 "$SKILL_DIR/scripts/extract_epub.py" "INPUT_PATH" "$WORK_DIR"
69	```
70
71	For PDF:
72	```bash
73	WORK_DIR=".book-work-$(date +%s)"
74	"$SKILL_DIR/scripts/extract_pdf.sh" "INPUT_PATH" "$WORK_DIR"
75	```
76
77	For MOBI: Convert to EPUB first with `ebook-convert` (calibre), then run EPUB extraction.
78
79	The work directory is created inside the vault root (e.g., `.book-work-1709398200/`).
80	Clean it up after the pipeline completes.
81
82	After extraction: Read `$WORK_DIR/metadata.json` to verify success.
83
84	PDF fallback: If metadata shows `"needs_fallback": true`, use the Claude Read tool instead:
85	- Read the PDF 20 pages at a time: `Read(file_path="INPUT_PATH", pages="1-20")`, then `pages="21-40"`, etc.
86	- Write each batch to `$WORK_DIR/pages/page_batch_NNN.txt`
87	- Update metadata with actual content
88
89	## Stage 2: STRUCTURE
90
91	Read `metadata.json` from the extraction output.
92
93	For EPUBs: Chapter structure comes directly from extraction (H1/H2 splitting). Proceed to Stage 3.
94
95	For PDFs: Chapter detection depends on extraction quality.
96	- If pages have clear chapter headings, group pages into chapters
97	- If no clear structure: Launch a single agent to read the first 5 pages and detect chapter boundaries:
98
99	```
100	Agent(
101	subagent_type="general-purpose",
102	model="haiku",
103	prompt="Read the following text from the first 5 pages of a book. Identify chapter
104	boundaries and return a JSON array of {title, start_page, end_page} objects.
105	If no chapters are detectable, return [{title: 'Full Text', start_page: 1, end_page: LAST}].
106
107	TEXT:
108	[first 5 pages content]"
109	)
110	```
111
112	- Update metadata with detected chapter list
113	- Group page files into chapter files by concatenation
114
115	Result: A `chapters/` directory with one file per chapter, and updated `metadata.json`.
116
117	## Stage 3: ANALYZE (parallel, file-based)
118
119	Read the agent definition from `agents/

People who liked this also viewed

6 picks

Type