reversa-archaeologist

To document undocumented legacy code, this Skill performs module-by-module reverse engineering to extract algorithms, control flow, and business logic directly from the source. It transforms the codebase into structured engineering artifacts like flowcharts and data dictionaries for downstream teams.

Extracting a data dictionary from an undocumented codebase involves analyzing domain data to derive entities and DTOs. This process documents field types, defaults, and relationships, outputting a structured data-dictionary.md file when configured to run at a detailed documentation level.

Generating module flowcharts from source code is achieved through deep reverse engineering analysis. By evaluating each planned module's control flow and validation logic, the system produces visual flowcharts based on your chosen documentation depth level.

Yes, you can resume reverse engineering a large legacy codebase because the process uses checkpointed excavation. It updates progress in a state.json file after each module, enabling resumable high-effort analysis without losing prior work.

Yes, a scout analysis is required before deep code excavation. The archaeologist workflow applies during the excavation phase after a scout, requiring existing state.json, plan.md, and context surface files to read and process the planned modules effectively.