PDF Document Assistant

Problem

PDF workflows often require small but frequent operations: combining files, splitting pages, extracting text, adding a password, and inspecting basic file metadata. These tasks can be scattered across desktop tools or ad hoc scripts, and dependency setup can slow down a restricted environment. PDF Document Assistant packages the common operations as callable Python functions, making it easier for an agent to handle file-level PDF tasks.

How It Works

The skill centers on utility functions in scripts/tool.py, preferring PyMuPDF(fitz) where available and attempting a fallback when external dependencies are limited. Core capabilities include:
- Merging PDFs: merge_pdfs(paths, output) combines multiple files into one output document.
- Splitting PDFs: split_pdf(path, output_dir) writes individual page files for distribution or review.
- Extracting text: extract_text(path) reads text content from a PDF for summarization or search.
- Encrypting PDFs: encrypt_pdf(path, password, output) applies an open password to a file.
- Analyzing metadata: analyze_pdf(path) returns basic information about the document.
Callers provide inputs such as file paths, output directory, password, and target path, and the functions return structured results that an agent can parse.

Boundaries

This skill is best suited to local or agent-mediated processing of existing PDF files: merging, splitting, text extraction, encryption, and basic analysis. It is not a hosted conversion service, an OCR pipeline, or a layout editor. For scanned documents, complex reflow, or batch server workloads, use dedicated tools. Before running, confirm that source paths are accessible, output directories are writable, and any protected PDF has the expected state.