Skip to content

espennilsen/pi / skills/pdf-reader

pdf-reader

Unverified

Read and extract content from PDF files — text, tables, metadata, and images. Use when asked to read a PDF, extract text from a PDF, summarize a PDF, analyze a PDF document, get tables from a PDF, or check PDF metadata. Also triggers on "open this PDF", "what does this PDF say", "parse PDF", "PDF to text", or when a .pdf file path or URL is provided.

PDF Reader

Extract content from PDF files using pdftotext (Poppler) for text and pdfplumber (Python) for tables and structured extraction.

Quick Reference

TaskToolCommand
Full textpdftotextpdftotext file.pdf -
Text with layoutpdftotextpdftotext -layout file.pdf -
Specific pagespdftotextpdftotext -f 3 -l 5 file.pdf -
Tablespdfplumberpython3 scripts/extract.py tables file.pdf
Metadatapdfinfopdfinfo file.pdf
Page countpdfinfopdfinfo file.pdf | grep Pages
Imagespdfimagespdfimages -list file.pdf
Fontspdffontspdffonts file.pdf
OCR (scanned PDF)tesseractpython3 scripts/extract.py ocr file.pdf
Smart text + OCRextract.pypython3 scripts/extract.py text file.pdf
Quick surveyextract.pypython3 scripts/extract.py scan file.pdf

Workflow

Step 1: Get the PDF

If the user provides a URL, download it first:

curl -sL "URL" -o /tmp/document.pdf

Verify it's a valid PDF:

file /tmp/document.pdf  # should say "PDF document"
pdfinfo /tmp/document.pdf  # metadata + page count

Step 2: Choose Extraction Method

Plain text (most cases):

pdftotext file.pdf -

This pipes output to stdout. For large PDFs, use page ranges:

pdftotext -f 1 -l 10 file.pdf -    # pages 1-10

Layout-preserving text (columns, formatted docs):

pdftotext -layout file.pdf -

Use -layout when the PDF has multi-column layouts, tables rendered as text, or precise spacing that matters.

Tables (structured data):

python3 scripts/extract.py tables file.pdf

Or inline with pdfplumber:

import pdfplumber

pdf = pdfplumber.open("file.pdf")
for i, page in enumerate(pdf.pages):
    tables = page.extract_tables()
    for table in tables:
        print(f"\n--- Table on page {i+1} ---")
        for row in table:
            print(" | ".join(str(cell or "") for cell in row))
pdf.close()

Metadata only:

pdfinfo file.pdf

Returns: title, author, creator, producer, page count, page size, dates.

Step 3: Handle Large PDFs

For PDFs over ~50 pages, don't dump everything at once:

  1. Get page count: pdfinfo file.pdf | grep Pages
  2. Extract in chunks: pdftotext -f 1 -l 20 file.pdf -
  3. Process chunk, then continue: pdftotext -f 21 -l 40 file.pdf -

For targeted extraction (searching for specific content):

# Extract all text, grep for relevant sections
pdftotext file.pdf - | grep -n -i "keyword"

# Then extract the specific page range
pdftotext -f PAGE -l PAGE file.pdf -

Step 4: Handle Scanned PDFs (OCR)

If pdftotext returns empty or garbled output, the PDF is likely scanned.

Detection:

python3 scripts/extract.py scan file.pdf   # reports scanned pages
pdffonts file.pdf                           # empty = image-based

Smart extraction (auto-fallback):

text mode automatically detects scanned pages and OCRs them:

python3 scripts/extract.py text file.pdf

Pages with selectable text extract normally. Pages without selectable text fall back to OCR via Tesseract. No manual detection needed.

Force OCR on all pages:

python3 scripts/extract.py ocr file.pdf
python3 scripts/extract.py ocr file.pdf --pages 1-5
python3 scripts/extract.py ocr file.pdf --dpi 400        # higher quality
python3 scripts/extract.py ocr file.pdf --lang eng+nor   # multi-language

OCR options:

  • --dpi 300 — resolution for page-to-image conversion (default: 300, higher = slower but better)
  • --lang eng — Tesseract language pack (default: eng). Use + for multiple: eng+nor+deu
  • --pages 1-5 — limit to specific pages (recommended for large PDFs)

Available language packs:

tesseract --list-langs

Install additional languages via Homebrew:

brew install tesseract-lang    # all languages

Step 5: Handle Other Edge Cases

Mixed PDFs (some pages scanned, some not):

Just use text mode — it handles mixed PDFs automatically:

python3 scripts/extract.py text file.pdf

Selectable pages extract instantly, scanned pages get OCR'd. The output is tagged so you know which pages used OCR.

Password-protected PDFs:

pdftotext -upw "password" file.pdf -   # user password
pdftotext -opw "password" file.pdf -   # owner password

Encoding issues (garbled output):

pdftotext -enc UTF-8 file.pdf -

Extract images:

pdfimages -png file.pdf /tmp/images/img   # extracts as PNG
pdfimages -list file.pdf                  # list images without extracting

Decision Tree

Is it a URL? → curl -sL "URL" -o /tmp/doc.pdf
            ↓
Run: python3 scripts/extract.py scan file.pdf
            ↓
All pages have selectable text?
  YES → pdftotext file.pdf -           (fast, simple)
  NO  → python3 scripts/extract.py text file.pdf  (auto OCR fallback)
            ↓
Need tables?
  YES → python3 scripts/extract.py tables file.pdf

Tips

  • Start with scan on unknown PDFs — it reports pages, tables, scanned detection, and a preview
  • pdftotext is fastest for normal PDFs — try it first
  • Use -layout for multi-column documents (academic papers, reports)
  • pdfplumber is better for tables — it understands cell boundaries
  • text mode auto-detects scanned pages and OCRs only those — preferred over raw pdftotext for unknown PDFs
  • ocr mode is for forcing OCR on everything (useful when text extraction gives garbled output despite appearing selectable)
  • Higher --dpi gives better OCR accuracy but is slower (300 is a good default, 400+ for small text)
  • For PDFs from URLs, always download to /tmp/ first — don't pipe curl to tools
  • Large PDF text output may exceed context limits — use --pages to extract in ranges
canvas-designanthropics/skillsCreate beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
docxanthropics/skillsUse this skill whenever the user wants to create, read, edit, or manipulate Word documents (.docx files) or Word templates (.dotx files). Triggers include: any mention of 'Word doc', 'word document', '.docx', '.dotx', or requests to produce professional documents with formatting like tables of contents, headings, page numbers, or letterheads. Also use when extracting or reorganizing content from .docx or .dotx files, inserting or replacing images in documents, performing find-and-replace in Word files, working with tracked changes or comments, or converting content into a polished Word document. If the user asks for a 'report', 'memo', 'letter', 'template', or similar deliverable as a Word or .docx file, use this skill. Do NOT use for PDFs, spreadsheets, Google Docs, or general coding tasks unrelated to document generation.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
pdfanthropics/skillsUse this skill whenever the user wants to do anything with PDF files. This includes reading or extracting text/tables from PDFs, combining or merging multiple PDFs into one, splitting PDFs apart, rotating pages, adding watermarks, creating new PDFs, filling PDF forms, encrypting/decrypting PDFs, extracting images, and OCR on scanned PDFs to make them searchable. If the user mentions a .pdf file or asks to produce one, use this skill.Stars: 196kUpdated Claude CodeCodexVerifiedInstall
pptxanthropics/skillsUse this skill any time a .pptx or .potx file is involved in any way — as input, output, or both. This includes: creating slide decks, pitch decks, or presentations; reading, parsing, or extracting text from any .pptx or .potx file (even if the extracted content will be used elsewhere, like in an email or summary); editing, modifying, or updating existing presentations; combining or splitting slide files; working with templates (.potx), layouts, speaker notes, or comments. Trigger whenever the user mentions "deck," "slides," "presentation," or references a .pptx or .potx filename, regardless of what they plan to do with the content afterward. If a .pptx or .potx file needs to be opened, created, or touched, use this skill.Stars: 196kUpdated Claude CodeCodexVerifiedInstall