@spoonly/mskr

Mskr

Instant image privacy masking from your terminal.

Installation

npm install -g @spoonly/mskr
# or using pnpm
pnpm install -g @spoonly/mskr

Usage

Mskr uses AI vision models to detect and mask sensitive information (names, phone numbers, etc.) in your images.

Quick Start

• Configure your API Key:

  mskr config set apiKey your_openai_api_key
  

• Mask an image:

  mskr input.jpg
  

  This will create input-masked.jpg by default. You can also use mskr start input.jpg.

Configuration

Manage your persistent settings to avoid repetitive typing.

# Interactive configuration menu
mskr config

# Direct set/get
mskr config set apiKey your_openai_api_key
mskr config set prompt "Find {targets} in the image and return coords [[x1,y1,x2,y2]]"
mskr config list

Commands

Command	Description
mskr [input]	Detects and masks sensitive info in the image (default).
mskr start [input]	Alias for masking (useful if filename conflicts with subcommands).
mskr config [action]	Manages configuration (apiKey, baseUrl, model, targets, prompt).

Masking Options

Option	Description
-o, --output <path>	Specify the output image path.
-t, --targets <list>	Custom targets to mask (e.g., "id card, address").
--prompt <template>	Custom prompt template (use {targets} placeholder).
--api-key <key>	Override stored API key.
--base-url <url>	Override stored Base URL.
--model <model>	Override stored Vision model.

Global Options

Option	Description
--debug	Enters debug mode, enabling detailed background logging utilizing consola.
-h, --help	Display full global help usage and command lists.
-v, --version	Display the currently installed version dynamically.

Note for Developers

This starter recommands using npm Trusted Publisher, where the release is done on CI to ensure the security of the packages.

To do so, you need to run pnpm publish manually for the very first time to create the package on npm, and then go to https://www.npmjs.com/package/@spoonly/mskr/access to set the connection to your GitHub repo.

Then for the future releases, you can run pnpm run release to do the release and the GitHub Actions will take care of the release process.

License

MIT License © OSpoon