Pdf Power

PDF toolkit: extract, merge, split, compress, convert, watermark, and protect PDFs.

Complete PDF toolkit. Extract text, merge documents, split pages, compress files, convert formats, add watermarks, and protect with passwords.

Dependencies

The skill uses Python with pypdf + PyMuPDF (fitz). Install on first use:

bash

pip install pypdf PyMuPDF Pillow

Operations

📄 Extract Text

Read all text from a PDF. Clean output, preserves paragraph structure.

🔗 Merge PDFs

Combine multiple PDFs into one file. Maintains page order.

✂️ Split PDF

Split a PDF by page ranges or extract specific pages.

📦 Compress PDF

Reduce file size. Offers two levels: light (fast, minimal loss) and heavy (smaller, may reduce quality).

🖼️ Convert PDF to Images

Convert each page to PNG/JPEG image. Useful for previews or presentations.

🖼️ Extract Images

Extract embedded images from PDF pages. Saves to output folder.

💧 Watermark

Add text or image watermark to every page. Configurable opacity and position.

🔒 Protect / Unprotect

Add or remove password protection on PDF files.

Usage Pattern

1. Identify the PDF path and desired operation.
2. Run the appropriate Python script.
3. Output file is saved beside the input file (or in specified output path).
4. Confirm result with the user.

Output

• Text extraction → clean text content
• Merge/Split/Compress/Protect → PDF file
• Convert to images → folder of image files
• Extract images → folder of image files
• Watermark → watermarked PDF

Critical Rules

• Never overwrite the original file unless user explicitly asks. Append _output to output filename.
• For large PDFs (>100 pages), warn the user and ask before proceeding.
• Validate PDF is not corrupted before processing.
• For password-protected PDFs, ask for the password upfront.
• Use pypdf for simple operations (merge, split, compress). Use PyMuPDF for rendering and image extraction.
• Compress only when user asks — don't compress silently.

---

⭐ Gostou desta skill? Deixe uma estrela no ClawHub para ajudar outros a encontrá-la!