dandy-code

Review AI-assisted code changes for continuation bias, missing context, architectural asymmetry, security gaps, and code ownership.

Review method and object APIs for argument count, option shape, boolean flags, object parameters, and fluent configuration.

Identify small, reusable components that can be isolated from a large codebase without prematurely forcing microservices.

Review error handling, logging, exception clarity, and debugging seams so failures stay visible and testable.

Simplify control flow with early returns, shallow nesting, explicit conditions, and readable branching.

Review whether code works with framework conventions instead of fighting them with unnecessary abstractions, mixed ecosystems, or upgrade-hostile choices.

Review and improve names for variables, methods, classes, files, routes, booleans, units, and paired operations.

Improve code readability by tightening formatting, visual grouping, method size, comments, and dead-code removal.

Review and improve repository README files so a new developer can understand, install, test, navigate, and find ownership quickly.

Design and review tests that drive clean code through unit focus, clear setup, independence, coverage awareness, and deterministic async checks.

Replace magic values, primitive obsession, nested arrays, fragile references, and scattered conversions with expressive objects or constants.