pdf-extract

name	pdf-extract
description	Extract schedules, text, and tables from PDF construction drawings. Can search for specific terms to find relevant pages. Results saved to output/ for cross-validation.
allowed-tools	Bash, Read, Write
argument-hint	[file.pdf] [command:text|tables|schedules|search] [options]

/pdf-extract — PDF Data Extraction

Extract data from PDF construction documents. For full CLI command syntax and flags, see /cli-tools.

Step 1: Find the PDF file

If no file path provided:

ls data/*.pdf 2>/dev/null

Multiple files → ask user. No files → tell user to place PDF in data/.

Step 2: Choose the right approach

• No specific command → start with schedules (auto-detects what's in the PDF)
• User asks about a topic → use search first to find pages, then tables on those pages
• User specifies pages → go directly to tables or text

Step 3: Run the extraction

Scripts are at skills/cli-tools/scripts/pdf_extract.py. Always include --save.

python skills/cli-tools/scripts/pdf_extract.py <file.pdf> <command> [options] --save

Common patterns:

# What schedules are in this PDF?
schedules --save

# Find the door schedule
search "door schedule" --save
# then: tables --pages <matched_pages> --save

# Extract everything from a page
text --pages 5 --save
tables --pages 5 --save

Step 4: Present results

• For schedules: list each type found, page, headers, row count
• For search: matching pages with snippets, suggest extraction commands
• For tables: describe each table — headers, rows, what it contains
• Suggest next steps: "Run /cross-validate door to compare against IFC model"

Error handling

• Empty tables → fall back to text extraction
• No schedules found → try search with specific terms
• Corrupted PDF → report structured error to user