Deprecated: These tools have been removed as built-in tools. Use bash commands with appropriate packages or install a skill that provides this functionality. The tools are documented here for reference only.
Text output is capped at 50,000 characters (MAX_TEXT_OUTPUT). Uses pdf-lib for creation/merging and pdftotext CLI for text extraction.
pdf_read — Read PDF Text
Extract text content from a PDF file.
Parameters
| Parameter | Type | Required | Description |
|---|
path | string | yes | Path to the PDF file |
pages | number[] | no | Specific pages to extract (1-indexed) |
max_chars | number | no | Maximum characters to return (default 50000) |
Returns
Extracted text content from the PDF, truncated at max_chars.
Notes
- Page numbers are 1-indexed
- Falls back to returning metadata only if
pdftotext CLI is not available on the system
- Text is extracted page by page with page markers
pdf_create — Create PDF
Create a new PDF file with text content.
Parameters
| Parameter | Type | Required | Description |
|---|
path | string | yes | Destination file path |
pages | array of {text, font_size?} | yes | Page content (min 1). Each object has a text string and an optional font_size number |
title | string | no | PDF document title metadata |
author | string | no | PDF document author metadata |
Returns
Confirmation with the file path and page count.
Notes
- Uses Helvetica font
- 50pt margins on all sides
- Text is auto-wrapped to fit within page margins
pdf_merge — Merge PDFs
Combine multiple PDF files into a single document.
Parameters
| Parameter | Type | Required | Description |
|---|
inputs | string[] | yes | Paths to input PDF files (min 2) |
output | string | yes | Destination path for the merged PDF |
Returns
Confirmation with the output path and total page count.
Notes
- PDFs are merged in the order provided
- All pages from each input file are included
Get metadata and structure information about a PDF file.
Parameters
| Parameter | Type | Required | Description |
|---|
path | string | yes | Path to the PDF file |
Returns
Object with page count, title, author, and page dimensions (width and height for the first 10 pages).
Notes
- Dimensions are reported for the first 10 pages only
- Returns available metadata fields (title, author, subject, creator, producer, creation date, modification date)
