pdf

Source: ComposioHQ/awesome-claude-skills Adapted for: LiYe OS Three-tier Architecture

When to Use This Skill

• Extract text, tables, or metadata from PDFs
• Create new PDF documents
• Merge or split PDF files
• Add annotations or comments
• Handle PDF forms

Core Capabilities

1. Text Extraction

• Extract full-text content
• Preserve document structure
• Identify headings and paragraph hierarchies
• OCR support (for scanned documents)

2. Table Extraction

• Recognize table structures
• Export as structured data
• Support complex nested tables

3. Metadata Processing

• Read document properties
• Extract information such as author, creation date, etc.
• Modify document metadata

4. Document Operations

• Merge multiple PDFs
• Split into multiple files
• Reorder pages
• Extract specific pages

5. Annotations & Forms

• Add highlights and comments
• Fill form fields
• Create fillable forms

Usage Examples

Example 1: Extract Research Paper Content

User: Extract the abstract and conclusion of this medical paper
Claude: [Use the pdf skill to extract content from the specified sections]

Example 2: Merge Multiple Reports

User: Merge these quarterly reports into an annual summary
Claude: [Use the pdf skill to merge files and add a table of contents page]

Example 3: Extract Table Data

User: Extract the data table from this PDF report
Claude: [Use the pdf skill to identify and extract the table, and convert it to a structured format]

Dependencies

• Python: PyPDF2, pdfplumber, reportlab, PyMuPDF
• Or Node.js: pdf-lib, pdf-parse

LiYe OS Integration

Business Domain References

• 05_Medical_Intelligence: Medical literature, clinical guideline parsing
• 01_Research_Intelligence: Academic paper analysis
• 04_Business_Operations: Business document processing

Three-tier Architecture Positioning

• Physical Layer (This file): Skills/00_Core_Utilities/document-processing/pdf/
• Logical Layer Index: Skills/{domain}/index.yaml
• L3 Instruction Layer: .claude/skills/{domain}/pdf/