reversa

Orchestrate legacy codebase reverse-engineering into agent-consumable specifications with resumable state.

1|Updated Dec 20, 2025
One-click install
npx skills add https://github.com/Rdinda/FMA_Pontos --skill reversa-rdinda
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reversa
Source: https://github.com/Rdinda/FMA_Pontos/tree/main/.agents/skills/reversa
Command: npx skills add https://github.com/Rdinda/FMA_Pontos --skill reversa-rdinda

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

It eliminates the time sink of understanding and documenting legacy projects by converting unclear existing codebases into structured, executable specifications for AI agents.

Core Features & Use Cases

  • Legacy system orchestration: coordinates a full reverse-engineering workflow starting from an initial project mapping and progressing through analysis, spec generation, and optional review.
  • Checkpointed, resumable execution: reads and writes state in .reversa/state.json to resume safely across sessions and phases without losing progress.
  • Configurable documentation output: enforces a documentation granularity choice (Essencial/Completo/Detalhado) and persists spec organization in .reversa/config.toml.
  • Agent-plan driven extraction: sequentially activates corresponding sub-skills (e.g., reversa-scout, reversa-archaeologist) and updates .reversa/plan.md and state checkpoints.

Quick Start

Use the command "/reversa" to start a complete legacy mapping and generate agent-ready specifications in the output folder.

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 codebase into AI agent-ready specifications?

Checkpointing in reverse engineering workflows saves progress atomically to a state.json file, allowing you to resume spec generation safely across sessions. This prevents data loss and avoids restarting the analysis if an extraction phase is interrupted.

Can I configure documentation granularity when generating specs for legacy systems?

Yes, you can configure documentation granularity for legacy systems by selecting Essential, Complete, or Detailed output levels. The chosen spec organization is persisted in a config.toml file, ensuring the generated agent documentation matches your project requirements.

What is the best way to document unclear legacy systems for multi-agent orchestration?

The best way to document unclear legacy systems for multi-agent orchestration is using a sequential agent execution plan. This approach maps the codebase, drives component extraction through specialized sub-skills, and updates a plan.md file to track spec generation progress.

Does this legacy system spec generation approach require any external dependencies?

No external dependencies are required for legacy system spec generation. The Skill operates independently by managing state and configuration locally through internal checkpointing and sequential sub-skill activation, requiring only your existing codebase to begin the analysis.