package-first-development

Package-first development identifies and reuses existing battle-tested libraries to minimize custom code. Search for packages matching your feature need, evaluate them on popularity, maintenance quality, and compatibility with your stack, then install the best fit—saving development time and reducing bugs.

Evaluate candidates on four criteria: popularity (download volume, GitHub stars), maintenance (recent updates, active issues), quality (documentation, TypeScript support), and compatibility with your framework and Node/Python version. This workflow prevents adopting abandoned or poorly-maintained libraries.

Package-first development works across npm and pip ecosystems. Specify your framework and language when requesting recommendations—the workflow searches for battle-tested libraries in your stack, evaluates cross-platform fit, and recommends options compatible with both your tech and project dependencies.

Instead of designing and building features from scratch, automated discovery uses web search to surface existing solutions, validates their quality and maintenance status, and presents ranked recommendations. This reduces feature-to-deployment time from weeks to hours for common tasks like form validation or image cropping.

Clarify your feature requirement, target framework (React, Vue, Django), and language ecosystem (npm, pip). Know your project's Node or Python version and any existing dependencies. This context ensures package recommendations align with your stack's compatibility and performance constraints.

Avoid external packages for highly specialized business logic, internal-only features with unique requirements, or when licensing restrictions apply. Custom code is better when no maintained package exists, when package overhead exceeds your use case, or when vendoring dependencies creates security risk.