summarize

Summarize large PDF, DOCX, TXT, and Markdown documents into executive and section reports.

6|Updated Feb 10, 2026
One-click install
npx skills add https://github.com/jdrodriguez/legal-toolkit --skill summarize-jdrodriguez
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: summarize
Source: https://github.com/jdrodriguez/legal-toolkit/tree/main/legal-toolkit/skills/summarize
Command: npx skills add https://github.com/jdrodriguez/legal-toolkit --skill summarize-jdrodriguez

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pdfplumber, python-docx, docx, PyMuPDF, tiktoken, and includes scripts (resource) components.

What problem does it solve?

Large documents and mixed-document directories are difficult to read in a single pass and time-consuming to synthesize into actionable insights. This Skill automates chunking, section detection, and parallel summarization so users quickly get an executive summary, section-by-section summaries, and a professional .docx report.

Core Features & Use Cases

  • Formats & inputs: Supports PDF, DOCX, TXT, and Markdown files as a single file or a directory of mixed documents.
  • Chunking & structure-aware processing: Detects section boundaries, splits text into token-aware chunks with overlap, and merges or splits sections as needed.
  • Parallel summarization: Coordinates multiple summarizer agents for medium/large jobs and collects section summaries into a unified report.
  • Output & workflows: Produces a final .docx report and a plain-text final_summary.md alongside chunk/summaries work files; includes dependency checks and graceful fallbacks for extraction.
  • Error handling: Handles unsupported formats, empty extraction (suggests OCR), missing scripts, and agent failures with clear user prompts.

Quick Start

Summarize the folder /path/to/docs into a unified executive summary and produce the final .docx report.

Frequently Asked Questions about summarize

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I summarize large PDF and DOCX files that are too long to read in one pass?

Large PDF and DOCX files are processed by splitting text into token-aware chunks with overlap, detecting section boundaries, and coordinating parallel summarizer agents to produce an executive summary and a .docx report.

Can I process a directory of mixed documents into a single executive summary?

Yes, you can process a directory of mixed documents into a single executive summary. The tool handles multi-file inputs including PDF, DOCX, TXT, and Markdown, merging section summaries into a unified report.

Do I need to install pdfplumber and python-docx to generate summary reports?

Yes, pdfplumber and python-docx are required dependencies for PDF text extraction and .docx report generation. The process includes dependency checks and prompts you if anything is missing.

What happens if text extraction fails on a scanned PDF during document summarization?

When text extraction fails on a scanned PDF during document summarization, the process handles the empty extraction gracefully and suggests using OCR to retrieve the text.

What is the best way to generate section-level reports for multiple large text files?

The best way to generate section-level reports for multiple large text files is using parallel agent coordination. Agents process token-aware chunks simultaneously, collecting section summaries into a final .docx report and plain-text summary.

Does chunking with overlap help when summarizing documents with complex section boundaries?

Chunking with token-aware overlap helps when summarizing documents with complex section boundaries by merging or splitting sections as needed, preserving context across chunks during parallel summarization.