Installation

Sluice is a Node.js CLI tool published to npm as @caracal-lynx/sluice.

Prerequisites

• Node.js 24 LTS or newer. Sluice depends on @duckdb/node-api, which targets the current Node ABI. Download from nodejs.org.
• A terminal. Sluice is a CLI; there is no UI. On Windows the team uses PowerShell 7 on Windows Terminal — anything POSIX or PowerShell-compatible works.

Confirm your Node version:

node --version
# → v24.x.x or newer

Install globally

The recommended install for day-to-day use:

npm install -g @caracal-lynx/sluice

Then verify the CLI is on your PATH:

sluice --version
sluice --help

You should see the package version and a list of commands (run, validate, profile, check, plugins, merge).

Install per-project

If you’d rather pin Sluice to a specific repo (recommended for client engagements where the version is locked alongside the pipeline YAML):

npm install --save-dev @caracal-lynx/sluice
npx sluice --version

In that mode every command becomes npx sluice <command> instead of plain sluice.

What you get

The @caracal-lynx/sluice npm package includes:

• The sluice CLI binary.
• Built-in source adapters for MSSQL, PostgreSQL, CSV, XLSX, and REST.
• Built-in target adapters for CSV and PostgreSQL.
• The full Data Quality and Transform engines.
• The plugin system (Tier 1 YAML composite rules, Tier 2 file plugins, Tier 3 npm packages).
• The DuckDB-backed staging layer.

ERP-specific target adapters (IFS, Business Central, BlueCherry), the Enrich service, country/region rule packages, and the MCP server are paid add-ons delivered via private npm packages — see Commercial Support.

Next step

You’re ready to run your first pipeline. Head to the Quickstart — under ten minutes from here to a working CSV-to-CSV migration with data quality validation.