skills.rest
pulak999
pulak999COMMUNITY
Explore all skills by this author →
GitHub

pdf-summarizer

Indexes and summarizes document folders into a persistent README.md cache.

Updated Mar 21, 2026
Search Tags:#summarization#pdf#knowledge base#text extraction#literature review#docx#document indexing
One-click install
npx skills add https://github.com/pulak999/ai-tooling --skill pdf-summarizer-pulak999
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: pdf-summarizer
Source: https://github.com/pulak999/ai-tooling/tree/main/ai-assistant-kit/cursor-skills/pdf-summarizer
Command: npx skills add https://github.com/pulak999/ai-tooling --skill pdf-summarizer-pulak999

SYSTEM DOCUMENTATION & REQUIREMENTS

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

What problem does it solve? Reading and re-reading large folders of PDFs, DOCX files, and notes for every question wastes time and context. This Skill builds a persistent README.md index of structured summaries so later questions are answered from the cache instead of re-opening every source file. ## Core Features & Use Cases - Folder-scale indexing: Scans a directory of PDF, DOCX, Markdown, and text files, then writes a structured five-point summary (problem, difficulty, rejected alternatives, solution, key details) per document into a README.md cache. - Staleness detection: Uses check-stale.sh to compare source modification times against summary dates and re-summarizes only STALE or MISSING entries. - README-first querying: Answers questions from the index and re-opens source files only for quotes, tables, or gaps, fetching targeted page ranges rather than whole documents. - Use Case: A researcher with 30 papers in a folder asks "what do these papers say about GPU virtualization" and gets an answer synthesized from the index without re-reading every PDF. ## Quick Start Ask the assistant to summarize and index all documents in a chosen folder so future questions about them are answered from the generated README index.

Frequently Asked Questions about pdf-summarizer

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

FAQPage Schema
How do I summarize a folder of PDF documents at once?▼

Point the skill at the folder and it lists all documents, extracts text with pdftotext or PyMuPDF, and writes a structured five-point summary per file into a README.md index. New files are batched in one pass and the README is written once at the end.

How to extract text from PDF files for summarization?▼

Use pdfinfo to inspect the file, sample pages with pdftotext, then run the extract-text.sh helper for full text. For PDFs over 40 pages, start with the abstract, introduction, and conclusion before reading methods or appendices.

Does PDF summarization work with scanned documents?▼

Scanned PDFs lack a text layer, so standard extraction returns nothing. The skill recommends sampling pages or installing OCR tools like tesseract-ocr or ocrmypdf when full text from scanned documents is required.

Can I index Word DOCX and Markdown files together with PDFs?▼

Yes, one README.md can index mixed folders containing PDFs, DOCX, Markdown, and plain text. DOCX files are read with python-docx, while Markdown and text files are read directly.

How are document summaries kept up to date when files change?▼

The check-stale.sh script compares each source file's modification time against its Summarized date in the README. Entries reported as STALE or MISSING are re-summarized in place, and entries are marked [STALE] until refreshed.

When should I not use folder-scale document indexing?▼

Avoid it for single-file coding reads or one-off questions about one document, where reading the file directly is faster. The index pays off only when a folder is queried repeatedly over time.

Related Skills

Explore More Education & Research Skills→
googleworkspacegoogleworkspace
30.3k

recipe-bulk-download-folder

List and download all files from a Google Drive folder, exporting Docs as PDFs.

Official
Basic
braselogbraselog
3

write-background

Draft manuscript background sections from literature summaries and citations.

Community
Basic
Guard8-aiGuard8-ai
6

sear

Convert PDFs and DOCX files to searchable Markdown with OCR.

Official
Advanced
dlantsdlants
7

pkb

Index local Markdown files and retrieve semantic search results with contextual information.

Community
Advanced