Sleek vs SQLFluff

Sleek is a command-line SQL formatter for applying consistent styles across SQL files. Its configurable options include indentation, keyword casing, and blank lines between queries. You can pass file paths or glob patterns, or provide input through standard input when no paths are supplied. The `--check` flag reports whether files are formatted without changing them. Sleek supports PostgreSQL, MySQL, MariaDB, SQLite, SQL Server, and Oracle. Its VS Code extension can format documents or selections, check formatting, and optionally format on save or paste; it can also download the CLI when it is not found. The CLI is available as compiled binaries or through Cargo, which requires Rust, and the project directs users to choose a binary for their operating system and architecture. Sleek is available under the MIT License, with formatting handled by the `sqlformat` crate. Building release v0.4.0 requires Rust 1.85 or later. The listed plan is free, with a free plan and no free trial.

SQLFluff is an open-source, dialect-flexible SQL linter designed for ELT applications. It reports style and syntax issues, parses SQL without database access and can automatically fix many violations with one command. Documentation describes support for more than 25 dialects, including T-SQL, PostgreSQL, BigQuery and Snowflake, alongside more than 70 rules for style, structure and best practices. Jinja templating and dbt are supported, and CI/CD integration includes GitHub Actions output for pull-request annotations. SQLFluff is free under the MIT License and installs from PyPI with pip or another Python package manager. It does not execute SQL itself, but templating macros such as dbt's run_query may execute arbitrary SQL. The Jinja templater uses Jinja2 SandboxedEnvironment to limit unsafe code execution. The dbt templater requires the relevant adapter and sqlfluff-templater-dbt package; models accessing a database during compilation require database access. Community support includes documentation, GitHub, Slack and Stack Overflow.