document-pro_skill

Overview

This skill gives an AI the ability to read, parse, extract, and summarize content from common document formats such as PDF, DOCX, PPTX, XLSX, TXT, and Markdown. It focuses on retrieving text, tables, and basic metadata so the AI can answer questions or produce concise reports. The skill helps convert formats and prepare structured outputs like summaries, key points, and CSV exports.

How this skill works

The skill detects the document type and selects appropriate parsers (pdfplumber/PyPDF2 for PDFs, python-docx for Word, python-pptx for presentations, openpyxl for spreadsheets). It extracts plain text, table data, and simple metadata, then runs lightweight analysis to identify structure, headings, and key points. For scanned or image-based PDFs, OCR is required before extraction; complex layouts or embedded charts may require manual review.

When to use it

• User uploads a document and asks for an analysis or summary
• Extract tables or data from reports, spreadsheets, or invoices
• Convert a document to or from PDF, Text, or Markdown
• Answer specific questions about content inside a document
• Prepare a concise report of key findings from a paper or presentation

Best practices

• Provide the original file when possible and indicate the expected output (summary, bullets, CSV)
• If document is a scanned image, supply OCR text or request OCR processing first
• Specify which sections or pages to prioritize if document is long
• For table extraction, confirm desired column mapping or CSV/Excel format
• Review extraction results for complex layouts, images, or charts before final use

Example use cases

• Summarize a 30-page research paper into a 1-page executive summary with 3–5 key takeaways
• Extract invoice line items and export them as CSV for bookkeeping
• Pull slide text from a PPTX and create speaker notes or a meeting brief
• Convert a DOCX policy document into Markdown and list major policy clauses
• Identify and extract tables from an annual report and save them to Excel