retro_requirements

You reverse-engineer requirements from a legacy codebase by applying a systematic 9-phase archaeology pipeline that identifies behaviors, extracts domain knowledge, and classifies confidence levels for documentation or migration planning.

The best way to extract system behaviors from undocumented source code is through systematic analysis that identifies what the system does rather than how it works, classifying extracted requirements by confidence levels for structured documentation.

Yes, you can analyze codebases written in any language or framework for migration planning because the reverse-engineering pipeline applies systematic analysis to extract domain knowledge regardless of the underlying technology stack.

Yes, you need Python and various analysis libraries installed to extract structured requirements from an application, as the reverse-engineering pipeline relies on these dependencies to execute its systematic 9-phase codebase analysis.

To resume an interrupted codebase analysis session, run the retro_requirements skill with the --continue flag, which restores the previous state of the systematic reverse-engineering pipeline.

You should use reverse engineering instead of forward requirements engineering when you have an existing, undocumented, or legacy codebase that needs structured descriptions for documentation, migration, or a from-scratch rebuild.