search-first

Evaluating existing libraries before coding prevents redundant development by enforcing a research-first workflow. Search package registries like npm and PyPI, as well as GitHub, to identify existing solutions and compare them against your custom code requirements.

Selecting dependencies for architectural decisions requires evaluating candidates based on maintenance, functionality, and licensing. Use a decision matrix framework to systematically assess whether to adopt, extend, or build custom solutions for your software engineering projects.

To find existing npm packages or MCP servers, use a systematic discovery process to search registries and GitHub repositories. This approach identifies production-proven tools, like using got instead of building a custom HTTP retry utility, before writing new code.

Build a custom solution instead of using an existing library when a decision matrix evaluation shows that available candidates fail to meet your project's required functionality, maintenance standards, or licensing constraints. Adopt, extend, or build based on this assessment.

Yes, you can apply the research-first workflow to Python PyPI package selection. The systematic discovery process searches across multiple registries including PyPI to evaluate existing solutions, ensuring patterns are assessed before custom code is written.