Shrinkr

Fast image compression & format conversion CLI tool. Batch compress JPG, PNG, TIFF, WebP, HEIC, AVIF, and GIF images with an interactive wizard or direct CLI flags.

Install

Homebrew (recommended)

brew install idrewlong/tap/shrinkr

Homebrew handles the libvips dependency automatically.

Go install

Requires Go and libvips installed on your system.

brew install vips       # macOS
go install github.com/idrewlong/shrinkr_cli@latest

From source

git clone https://github.com/idrewlong/shrinkr_cli
cd shrinkr_cli
make install

Usage

Interactive wizard

Run shrinkr with no arguments to launch the guided wizard:

shrinkr

The wizard will:

1. Auto-detect nearby folders with images — or browse with Finder / enter a path manually
2. Ask for output format (WebP, PNG, JPEG, or AVIF)
3. Ask for compression settings (Recommended, Web Optimized, High Quality, or Custom)
4. Ask where to save the output — defaults to a compressed/ folder inside your selected input folder
5. Show a summary and confirm before running

Tip: Press Esc or Ctrl+C at any step to go back one step. Pressing either key repeatedly from the first step exits the program.

Direct CLI mode

Pass the input folder as an argument for scripting or power use:

# Compress a folder to WebP at 500 KB target (default)
shrinkr ./photos

# Custom format, size, and output location
shrinkr ./photos -f jpeg -s 300 -o ./compressed-photos

# Recursive scan with quality settings
shrinkr ~/Pictures -r -f webp -s 200 -q 80

# All options
shrinkr <folder> [flags]

Flags

Flag	Short	Default	Description
--format	-f	webp	Output format: webp, png, jpeg, avif
--size	-s	500	Target file size in KB
--output	-o	compressed	Output folder path
--quality	-q	85	Initial quality (1–100)
--workers	-w	CPU count	Concurrent workers (auto-detected)
--recursive	-r	false	Scan subfolders
--min-quality		60	Quality floor
--max-quality		90	Quality ceiling

Supported formats

Input	Output
JPG / JPEG	WebP
PNG	PNG
TIFF / TIF	JPEG
WebP	AVIF
GIF
AVIF
HEIF / HEIC

AVIF note: AVIF produces the smallest files but encodes significantly slower than other formats — best for small batches where file size is the top priority.

How it works

Shrinkr uses a binary search algorithm to find the optimal compression quality that hits your target file size:

1. Tries initial quality (default 85)
2. If the result is over target, binary searches between min and max quality
3. Falls back to min quality if the target can't be reached
4. Runs all compressions concurrently using a worker pool sized to your CPU core count

Processing uses libvips via the bimg Go wrapper — one of the fastest image processing libraries available.

Development

# Run all tests
make test

# Run format benchmarks (shows per-format encoding speed comparison)
make bench

# Build binary locally
make build

# Build and install globally
make install

Updates

brew upgrade shrinkr

Requirements

• macOS (Apple Silicon or Intel)
• libvips — installed automatically by Homebrew

License

MIT