🔒 Metadata Scrubber

A privacy-focused CLI tool that removes sensitive metadata from files. Supports images, PDFs, and Microsoft Office documents. Perfect for protecting your privacy before sharing files online.

✨ Features

• Multi-format support - Images (JPEG, PNG), PDFs, and Office docs (Word, Excel, PowerPoint)
• Concurrent processing - Process 1000+ files efficiently with ThreadPoolExecutor
• Dry-run mode - Preview what would be scrubbed without making changes
• Verification reports - Before/after comparison to confirm removal
• Smart format detection - Uses library-level format detection, not just file extensions
• Beautiful CLI - Rich progress bars and formatted output
• Privacy-first - Removes GPS coordinates, author info, timestamps, camera data

📁 Supported Formats

Category	Extensions	Metadata Removed
Images	.jpg, .jpeg, .png	EXIF, GPS, camera info, timestamps
PDF	.pdf	Author, creator, producer, dates
Word	.docx	Author, title, comments, keywords
Excel	.xlsx, .xlsm, .xltx, .xltm	Author, title, company, comments
PowerPoint	.pptx, .pptm, .potx, .potm	Author, title, comments, keywords

🚀 Quick Start

Installation

# Using uv (recommended)
uv pip install metadata-scrubber

# Or clone and install locally
git clone https://github.com/Heritage-XioN/metadata-scrubber-tool.git
cd metadata-scrubber-tool
uv sync

Basic Usage

# Read metadata from a file
mst read document.pdf

# Scrub metadata and save to output folder
mst scrub photo.jpg --output ./cleaned

# Batch process entire folder
mst scrub ./documents -r -ext docx --output ./cleaned

# Verify removal
mst verify original.jpg ./cleaned/processed_original.jpg

📖 Commands

mst read - View Metadata

Extract and display all embedded metadata from a file.

mst read photo.jpg                      # Single file
mst read report.pdf                     # PDF file
mst read ./docs -r -ext docx            # All Word docs recursively

Example output:

╭────────────────── Metadata Report ──────────────────╮
│ ╭────────────────────┬────────────────────────────╮ │
│ │ Property           │ Value                      │ │
│ ├────────────────────┼────────────────────────────┤ │
│ │ 📷 Camera          │                            │ │
│ │   Make             │ Canon                      │ │
│ │   Model            │ Canon EOS 80D              │ │
│ │   Software         │ Adobe Photoshop            │ │
│ ├────────────────────┼────────────────────────────┤ │
│ │ 📍 GPS             │                            │ │
│ │   GPSLatitude      │ 40.7128                    │ │
│ │   GPSLongitude     │ -74.0060                   │ │
│ ├────────────────────┼────────────────────────────┤ │
│ │ 📅 Dates           │                            │ │
│ │   DateTimeOriginal │ 2024:01:15 14:30:00        │ │
│ │   created          │ 2024-01-15 14:30:00        │ │
│ ╰────────────────────┴────────────────────────────╯ │
╰─────────────────────────────────────────────────────╯

---

mst scrub - Remove Metadata

Remove sensitive metadata from files and save cleaned copies.

mst scrub photo.jpg --output ./out      # Single file
mst scrub ./photos -r -ext jpg -o ./out # All JPEGs in directory
mst scrub ./docs -r -ext pdf --dry-run  # Preview without changes
mst scrub ./files -r -ext xlsx -w 8     # 8 concurrent workers

Example output:

Processing 42 files with 4 workers...

⠸ Scrubbing metadata... ━━━━━━━━━━━━━━━━━━━━━━━━━━ 100% 42/42 0:00:12

╭───────────────────── Summary ─────────────────────╮
│ ✅ Processed: 42                                  │
│ ❌ Failed:    0                                   │
│ 📁 Output:    C:\Users\...\cleaned                │
╰───────────────────────────────────────────────────╯

Dry-run example:

mst scrub ./photos -r -ext jpg --dry-run

🔍 DRY-RUN MODE - No files will be modified

Would process 15 files:
  • photo1.jpg → processed_photo1.jpg
  • photo2.jpg → processed_photo2.jpg
  • vacation/beach.jpg → processed_beach.jpg
  ...

---

mst verify - Verify Metadata Removal

Compare original and processed files to confirm sensitive data was removed.

mst verify original.jpg ./out/processed_original.jpg

Example output:

Comparing: test_canon.jpg → processed_test_canon.jpg

                          Verification Report                          
┏━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━━━━━━━━━━━┳━━━━━━━━━━━━━━━━┓
┃ Property                ┃ Before                   ┃ After          ┃
┡━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━━━━━━━━━━━╇━━━━━━━━━━━━━━━━┩
│ Make                    │ Canon                    │ ✅ Removed     │
│ Model                   │ Canon EOS 80D            │ ✅ Removed     │
│ Software                │ Adobe Photoshop          │ ✅ Removed     │
│ GPSLatitude             │ 40.7128                  │ ✅ Removed     │
│ GPSLongitude            │ -74.0060                 │ ✅ Removed     │
│ Artist                  │ John Smith               │ ✅ Removed     │
│ Copyright               │ © 2024 John Smith        │ ✅ Removed     │
│ DateTimeOriginal        │ 2024:01:15 14:30:00      │ ⚪ Preserved   │
└─────────────────────────┴──────────────────────────┴────────────────┘

✅ Status: CLEAN - All sensitive metadata removed
Removed: 38 | Preserved: 2

---

⚙️ CLI Options

Option	Description
-r, --recursive	Process directories recursively
-ext, --extension	Filter by file extension (jpg, png, pdf, docx, xlsx, pptx)
-o, --output	Output directory for cleaned files
-d, --dry-run	Preview without making changes
-w, --workers	Number of concurrent workers (default: 4, max: 16)
-V, --verbose	Show detailed debug logs
-v, --version	Show version

---

🛠️ Development

Setup

git clone https://github.com/Heritage-XioN/metadata-scrubber-tool.git
cd metadata-scrubber-tool

# Install with dev dependencies
uv sync --all-extras

# Run tests
pytest

# Run linting
ruff check .

# Run type checking
mypy src

Project Structure

src/
├── main.py                   # CLI entry point (Typer app)
├── commands/
│   ├── read.py               # Read metadata command
│   ├── scrub.py              # Scrub metadata command
│   └── verify.py             # Verify removal command
├── services/
│   ├── metadata_factory.py   # Factory for creating handlers
│   ├── metadata_handler.py   # Abstract base class
│   ├── image_handler.py      # JPEG/PNG handler
│   ├── pdf_handler.py        # PDF handler
│   ├── excel_handler.py      # Excel handler
│   ├── powerpoint_handler.py # PowerPoint handler
│   ├── worddoc_handler.py    # Word document handler
│   ├── report_generator.py   # Verification reports
│   └── batch_processor.py    # Concurrent batch processing
└── core/
    ├── jpeg_metadata.py      # JPEG EXIF processor
    └── png_metadata.py       # PNG metadata processor

docs/
├── metadata-risks.md         # Privacy risks documentation
└── best-practices.md         # Secure file sharing guide

---

📚 Documentation

• Metadata Risks - Why metadata matters for privacy
• Best Practices - Guidelines for secure file sharing

---

⚠️ Known Limitations

File Format Support

Category	Supported	Not Supported
Images	JPEG, PNG	TIFF, GIF, HEIC, WebP, RAW
Documents	.docx	Legacy .doc
Spreadsheets	.xlsx, .xlsm, .xltx, .xltm	Legacy .xls
Presentations	.pptx, .pptm, .potx, .potm	Legacy .ppt
PDF	Standard PDFs	Encrypted/password-protected

Known Constraints

• No in-place editing - Always creates a processed copy (by design for safety)
• Password-protected files - Cannot process encrypted documents
• PNG metadata - Many PNGs have minimal/no extractable metadata
• Embedded files - Objects embedded in Office documents are not deep-scanned
• PDF embedded images - Images inside PDFs retain their original metadata
• Large files - Files are loaded into memory; very large files may be slow

PNG Verification Behavior

When a PNG file has no EXIF metadata (only PngInfo text chunks), the scrub operation removes all text keys. Attempting to verify or read the processed file will show:

Error during verification: No metadata found in the PNG image.

This is expected behavior - the error confirms that all metadata has been successfully removed. You can also use mst read processed_file.png to verify; the same error indicates a clean file.

Future Enhancements

• HEIC/HEIF support (common on iOS devices)
• Legacy Office format support (.doc, .xls, .ppt)
• Deep scanning of embedded objects
• PDF embedded image metadata stripping

---

⚠️ Security Considerations

• Original files are never modified - processed copies are created
• Use --dry-run to preview changes before committing
• Use mst verify to confirm sensitive data was removed
• GPS coordinates are completely stripped for privacy
• Author information is removed from all supported formats
• Always backup files before scrubbing in production

---

📄 License

MIT License - See LICENSE for details.

---

Made with ❤️ for privacy