Verified:GitHub

reversa

Orchestrate legacy system reverse-engineering into executable AI-ready specifications.

Updated May 5, 2026

Search Tags:#specification #checkpointing #reverse-engineering #ai-workflow #agent-orchestration #documentation-automation #legacy-system

Ready to run

npx skills add https://github.com/AlexandrePontesjr/calculadora-Juridica --skill reversa-alexandrepontesjr

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: reversa
Source: https://github.com/AlexandrePontesjr/calculadora-Juridica/tree/main/.agents/skills/reversa
Command: npx skills add https://github.com/AlexandrePontesjr/calculadora-Juridica --skill reversa-alexandrepontesjr

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

The Reversa skill turns a legacy system into actionable, executable specifications by orchestrating a multi-step analysis process, preserving progress across sessions.

Core Features & Use Cases

Legacy system mapping and plan execution: reads prior state and runs a sequential agent plan starting from the project “Scout” to map modules and integrations.
Stateful, checkpointed workflow: uses .reversa/state.json and step references to safely resume between phases without losing progress.
Documentation-level and spec-organization control: prompts for doc_level and persists the specs organization layout in .reversa/config.toml before generating deeper artifacts.

Quick Start

Activate the Reversa orchestrator by typing: /reversa.

Frequently Asked Questions about reversa

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I reverse-engineer a legacy system into executable AI specifications?▼

You orchestrate legacy system reverse-engineering into executable AI-ready specifications by running a multi-phase project analysis workflow that maps modules, applies checkpointing, and generates structured specs for downstream agents.

What is the best way to maintain progress when analyzing a large legacy codebase across sessions?▼

Maintaining progress when analyzing a legacy codebase across sessions requires checkpointing state in a dedicated JSON file, which preserves the sequential plan execution and safely resumes the workflow without losing prior module mapping data.

Can I control the documentation level and spec organization layout during reverse-engineering?▼

Yes, you can control the documentation level and spec organization layout by prompting for user approvals before generating deeper artifacts, then persisting the chosen configuration in a TOML file within the project state directory.

How does checkpointing work in an AI workflow for legacy system mapping?▼

Checkpointing in an AI workflow for legacy system mapping works by reading and updating a central state JSON file, enforcing sequential plan execution, and gating progress on user approvals to ensure documentation phases complete safely.

Where are generated specifications and reverse-engineering artifacts written during the workflow?▼

Generated specifications and reverse-engineering artifacts are written exclusively to allowed directories, including the state directory, the SDD output folder, and limited forward history paths to prevent unauthorized file modifications.