What problem does it solve? Reverse-engineering a legacy codebase requires deep, structured analysis that is hard to do consistently by hand. This Skill performs the excavation phase of a reverse-engineering workflow, reading each module of a legacy project and producing consolidated technical documentation with confidence levels for every finding. ## Core Features & Use Cases - Module-by-module analysis: Extracts control flow, non-trivial algorithms, data structures, entities, business rules, and configuration metadata from each module listed in the analysis plan. - Tiered documentation output: Generates code-analysis.md, data-dictionary.md, and Mermaid flowcharts scaled to the configured doc_level (essencial, completo, detalhado). - Structured machine-readable output: Produces modules.json following a defined schema with confidence levels (confirmed, inferred, unknown) consumed by downstream reverse-engineering agents. - Checkpointing and session pacing: Saves per-module checkpoints to state.json and offers preventive pauses after intensive analysis to preserve quality across sessions. - Use Case: After running the reversa-scout surface scan on an undocumented legacy system, invoke this Skill to excavate each module and produce a data dictionary and flowcharts before writing the new specification. ## Quick Start Ask the agent to run the reversa-archaeologist excavation on the modules listed in .reversa/plan.md and generate the code analysis and data dictionary.
