pdf

Name: pdf
Author: stvlynn

118stars21forks

How to install pdf?

stvlynn/dingtalk-wukong-skills/pdf

$npx -y skills add stvlynn/dingtalk-wukong-skills --skill pdf

Installs into the current project.

Summary

Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms. When Claude needs to fill in a PDF form or programmatically process, generate, or analyze PDF documents at scale.

Files · 1

View on GitHub

SKILL.md

1	# PDF文档处理专家 Real Skill
2
3	## 一句话说明
4
5	全面的 PDF 文档处理工具，支持文本表格提取、PDF 创建、文档操作和表单填写。
6
7	PDF Processing Guide - Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms.
8
9	---
10
11	## 使用场景
12
13	适用：
14	- 从 PDF 提取文本、表格数据并导出 Excel
15	- 创建新的 PDF 文档（报告、证书、发票）
16	- 合并、拆分、旋转 PDF 文档
17	- 填写 PDF 表单（政府表单、申请表）
18	- 批量处理多个 PDF 文件
19
20	不适用：
21	- 扫描版 PDF（需先 OCR 处理 / requires OCR first）
22	- 复杂的 PDF 编辑（用专业软件 / use professional software）
23
24	---
25
26	## 核心流程
27
28	```
29	用户需求 → 判断任务类型 → 选择工具 → 执行操作 → 验证输出
30	User request → Task type classification → Tool selection → Execute → Validate
31	```
32
33	任务类型判断 (Task Classification)：
34
35	### Extracting Data (提取数据)
36	- 文本提取 (Text extraction) → pdfplumber.extract_text()
37	- Better accuracy than pypdf for text extraction
38	- 表格提取 (Table extraction) → pdfplumber.extract_tables() + pandas
39	- Intelligent table structure recognition
40	- Export to Excel, CSV, JSON
41
42	### Creating PDFs (创建 PDF)
43	- 简单文档 (Simple documents) → reportlab.pdfgen.canvas
44	- Quick document creation with text and graphics
45	- 复杂布局 (Complex layouts) → reportlab.platypus
46	- Paragraphs, Tables, Images with advanced styling
47
48	### PDF Operations (PDF 操作)
49	- 合并 (Merge) → pypdf.PdfWriter
50	- Combine multiple PDFs into one
51	- 拆分 (Split) → pypdf.PdfWriter (save each page separately)
52	- 旋转 (Rotate) → page.rotate()
53	- 元信息 (Metadata) → pypdf.PdfReader.metadata / PdfWriter.add_metadata()
54
55	### Form Processing (表单处理)
56	- 查看字段 (View fields) → See forms.md
57	- 填写表单 (Fill forms) → See forms.md
58	- 批量处理 (Batch processing) → Loop + form filling API
59
60	---
61
62	## 任务完成标准
63
64	必须满足（缺一不可）：
65	- PDF 文件可正常打开
66	- 提取的数据完整准确
67	- 表格结构保留正确（行列对应）
68	- 创建的 PDF 内容清晰可读
69
70	质量评级：
71	- ⭐⭐⭐⭐⭐ 优秀 - 数据完整 + 格式规范 + 自动化批处理
72	- ⭐⭐⭐ 及格 - 数据正确 + PDF 可用
73	- ⭐ 失败 - 数据缺失或 PDF 损坏
74
75	---
76
77	## 参考资料（供 AI 使用）
78
79	\| 类型 \| 路径 \| 说明 \|
80	\|-----\|------\|------\|
81	\| 核心文档 \| `docs/00-SKILL-完整处理指南.md` \| Complete PDF processing guide (~300 lines) \|
82	\| 表单处理 \| `docs/forms.md` \| PDF form filling detailed tutorial \|
83	\| 高级参考 \| `docs/reference.md` \| JavaScript libraries and advanced features \|
84
85	---
86
87	## 关键原则（AI 必读 / Critical Principles）
88
89	### 1. Prefer pdfplumber for Extraction (优先使用 pdfplumber)
90	- 文本提取: More accurate than pypdf
91	- 表格识别: Strong table recognition capability
92	- 布局保留: Preserves layout information
93	- 复杂 PDF: Better for complex PDF processing
94
95	### 2. Table Data Normalization (表格数据规范化)
96	```python
97	# Best practice for table extraction
98	import pdfplumber
99	import pandas as pd
100
101	all_tables = []
102	with pdfplumber.open("document.pdf") as pdf:
103	for page in pdf.pages:
104	tables = page.extract_tables()
105	for table in tables:
106	if table: # Check not empty
107	df = pd.DataFrame(table[1:], columns=table[0]) # First row as header
108	all_tables.append(df)
109
110	# Combine and clean
111	combined = pd.concat(all_tables, ignore_index=True)
112	combined.to_excel("output.xlsx", index=False)
113	```
114
115	### 3. Scanned PDFs Require OCR (扫描版需要 OCR)
116	- 识别: If text extraction returns empty or garbled text
117	- 处理: Must use OCR tools first (Tesseract, Adobe Acrobat, etc.)
118	- 告知用户: Clearly inform user that OCR pre-processing is needed
119
120	### 4. Memory Management for Large Files (大文件内存管理)
121	- Batch processing for multiple PDFs
122	- Process pages sequentially rather than loading entire document
123	- Use generators when possible
124
125	---
126
127	## 快速命令参考 (Quick Commands)
128
129	### Text Extraction (文本提取)
130	```python
131	import pdfplumber
132
133	with pdfplumber.open("document.pdf") as pdf:
134	for page in pdf.pages:
135	text = page.extract_text()
136	print(text)
137	```
138
139	### Table to Excel (表格转 Excel)
140	```python
141	import pdfplumber
142	import pandas as pd
143
144	all_tables = []
145	with pdfplumber.open("document.pdf") as pdf:
146	for page in pdf.pages:
147	tables = page.extract_tables()
148	for table in tables:
149	if table:
150	df = pd.DataFrame(table[1:], columns=table[0])
151	all_tables.append(df)
152
153	combined = pd.concat(all_tables, ignore_index=True)
154	combined.to_excel("extracted_tables.xlsx", index=False)
155	```
156
157	### Merge PDFs (合并 PDF)
158	```python
159	from pypdf import PdfWriter, PdfReader
160
161	writer = PdfWriter()
162	for pdf_file in ["doc1.pdf", "doc2.pdf", "doc3.pdf"]:
163	reader = PdfReader(pdf_file)
164	for page in reader.pages:
165	writer.add_page(page)
166
167	with open("merged.pdf", "wb") as output:
168	writer.write(output)
169	```
170
171	### Split PDF (拆分 PDF)
172	```python
173	from pypdf import PdfReader, PdfWriter
174
175	reader = PdfReader("large_document.pdf")
176	for i, page in enumerate(reader.pages):
177	writer = PdfWriter()
178	writer.add_page(page)
179	with open(f"page_{i+1}.pdf", "wb") as output:
180	writer.write(output)
181	```
182
183	### Create PDF (创建 PDF)
184	``

People who liked this also viewed

6 picks

Type

pdf

118stars21forks

View on GitHub

How to install pdf?

stvlynn/dingtalk-wukong-skills/pdf

$npx -y skills add stvlynn/dingtalk-wukong-skills --skill pdf

Installs into the current project.

Summary

Files · 1

View on GitHub

SKILL.md

1	# PDF文档处理专家 Real Skill
2
3	## 一句话说明
4
5	全面的 PDF 文档处理工具，支持文本表格提取、PDF 创建、文档操作和表单填写。
6
7	PDF Processing Guide - Comprehensive PDF manipulation toolkit for extracting text and tables, creating new PDFs, merging/splitting documents, and handling forms.
8
9	---
10
11	## 使用场景
12
13	适用：
14	- 从 PDF 提取文本、表格数据并导出 Excel
15	- 创建新的 PDF 文档（报告、证书、发票）
16	- 合并、拆分、旋转 PDF 文档
17	- 填写 PDF 表单（政府表单、申请表）
18	- 批量处理多个 PDF 文件
19
20	不适用：
21	- 扫描版 PDF（需先 OCR 处理 / requires OCR first）
22	- 复杂的 PDF 编辑（用专业软件 / use professional software）
23
24	---
25
26	## 核心流程
27
28	```
29	用户需求 → 判断任务类型 → 选择工具 → 执行操作 → 验证输出
30	User request → Task type classification → Tool selection → Execute → Validate
31	```
32
33	任务类型判断 (Task Classification)：
34
35	### Extracting Data (提取数据)
36	- 文本提取 (Text extraction) → pdfplumber.extract_text()
37	- Better accuracy than pypdf for text extraction
38	- 表格提取 (Table extraction) → pdfplumber.extract_tables() + pandas
39	- Intelligent table structure recognition
40	- Export to Excel, CSV, JSON
41
42	### Creating PDFs (创建 PDF)
43	- 简单文档 (Simple documents) → reportlab.pdfgen.canvas
44	- Quick document creation with text and graphics
45	- 复杂布局 (Complex layouts) → reportlab.platypus
46	- Paragraphs, Tables, Images with advanced styling
47
48	### PDF Operations (PDF 操作)
49	- 合并 (Merge) → pypdf.PdfWriter
50	- Combine multiple PDFs into one
51	- 拆分 (Split) → pypdf.PdfWriter (save each page separately)
52	- 旋转 (Rotate) → page.rotate()
53	- 元信息 (Metadata) → pypdf.PdfReader.metadata / PdfWriter.add_metadata()
54
55	### Form Processing (表单处理)
56	- 查看字段 (View fields) → See forms.md
57	- 填写表单 (Fill forms) → See forms.md
58	- 批量处理 (Batch processing) → Loop + form filling API
59
60	---
61
62	## 任务完成标准
63
64	必须满足（缺一不可）：
65	- PDF 文件可正常打开
66	- 提取的数据完整准确
67	- 表格结构保留正确（行列对应）
68	- 创建的 PDF 内容清晰可读
69
70	质量评级：
71	- ⭐⭐⭐⭐⭐ 优秀 - 数据完整 + 格式规范 + 自动化批处理
72	- ⭐⭐⭐ 及格 - 数据正确 + PDF 可用
73	- ⭐ 失败 - 数据缺失或 PDF 损坏
74
75	---
76
77	## 参考资料（供 AI 使用）
78
79	\| 类型 \| 路径 \| 说明 \|
80	\|-----\|------\|------\|
81	\| 核心文档 \| `docs/00-SKILL-完整处理指南.md` \| Complete PDF processing guide (~300 lines) \|
82	\| 表单处理 \| `docs/forms.md` \| PDF form filling detailed tutorial \|
83	\| 高级参考 \| `docs/reference.md` \| JavaScript libraries and advanced features \|
84
85	---
86
87	## 关键原则（AI 必读 / Critical Principles）
88
89	### 1. Prefer pdfplumber for Extraction (优先使用 pdfplumber)
90	- 文本提取: More accurate than pypdf
91	- 表格识别: Strong table recognition capability
92	- 布局保留: Preserves layout information
93	- 复杂 PDF: Better for complex PDF processing
94
95	### 2. Table Data Normalization (表格数据规范化)
96	```python
97	# Best practice for table extraction
98	import pdfplumber
99	import pandas as pd
100
101	all_tables = []
102	with pdfplumber.open("document.pdf") as pdf:
103	for page in pdf.pages:
104	tables = page.extract_tables()
105	for table in tables:
106	if table: # Check not empty
107	df = pd.DataFrame(table[1:], columns=table[0]) # First row as header
108	all_tables.append(df)
109
110	# Combine and clean
111	combined = pd.concat(all_tables, ignore_index=True)
112	combined.to_excel("output.xlsx", index=False)
113	```
114
115	### 3. Scanned PDFs Require OCR (扫描版需要 OCR)
116	- 识别: If text extraction returns empty or garbled text
117	- 处理: Must use OCR tools first (Tesseract, Adobe Acrobat, etc.)
118	- 告知用户: Clearly inform user that OCR pre-processing is needed
119
120	### 4. Memory Management for Large Files (大文件内存管理)
121	- Batch processing for multiple PDFs
122	- Process pages sequentially rather than loading entire document
123	- Use generators when possible
124
125	---
126
127	## 快速命令参考 (Quick Commands)
128
129	### Text Extraction (文本提取)
130	```python
131	import pdfplumber
132
133	with pdfplumber.open("document.pdf") as pdf:
134	for page in pdf.pages:
135	text = page.extract_text()
136	print(text)
137	```
138
139	### Table to Excel (表格转 Excel)
140	```python
141	import pdfplumber
142	import pandas as pd
143
144	all_tables = []
145	with pdfplumber.open("document.pdf") as pdf:
146	for page in pdf.pages:
147	tables = page.extract_tables()
148	for table in tables:
149	if table:
150	df = pd.DataFrame(table[1:], columns=table[0])
151	all_tables.append(df)
152
153	combined = pd.concat(all_tables, ignore_index=True)
154	combined.to_excel("extracted_tables.xlsx", index=False)
155	```
156
157	### Merge PDFs (合并 PDF)
158	```python
159	from pypdf import PdfWriter, PdfReader
160
161	writer = PdfWriter()
162	for pdf_file in ["doc1.pdf", "doc2.pdf", "doc3.pdf"]:
163	reader = PdfReader(pdf_file)
164	for page in reader.pages:
165	writer.add_page(page)
166
167	with open("merged.pdf", "wb") as output:
168	writer.write(output)
169	```
170
171	### Split PDF (拆分 PDF)
172	```python
173	from pypdf import PdfReader, PdfWriter
174
175	reader = PdfReader("large_document.pdf")
176	for i, page in enumerate(reader.pages):
177	writer = PdfWriter()
178	writer.add_page(page)
179	with open(f"page_{i+1}.pdf", "wb") as output:
180	writer.write(output)
181	```
182
183	### Create PDF (创建 PDF)
184	``

People who liked this also viewed

6 picks

Type