reversa-reviewer

Reviews reverse-engineering specs and outputs confidence reports and questions files.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It reduces the risk of incorrect or incomplete reverse-engineering specifications by systematically challenging generated specs, identifying inconsistencies, and surfacing human-validation questions.

Core Features & Use Cases

  • Unit-by-unit spec review: Checks each generated unit for missing canonical files, internal consistency, and business-rule coherence.
  • Cross-unit contradiction detection: Looks for conflicting assumptions and missing modules across the full generated output set.
  • Confidence and gap reporting: Produces confidence summaries and, when needed, a structured questions file for unresolved critical gaps.

Quick Start

Run the reversa-reviewer workflow to review the units in the generated output folder and produce a confidence report plus a questions list for any blocking 🔴 gaps.

Frequently Asked Questions about reversa-reviewer

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

FAQPage Schema
How do I validate reverse-engineering specifications for missing files and inconsistencies?

To validate reverse-engineering specs, you review each generated unit for missing canonical files and internal consistency. This process detects cross-unit contradictions and reclassifies confidence levels across the full output set.

What is cross-unit contradiction detection in reverse engineering?

Cross-unit contradiction detection identifies conflicting assumptions and missing modules across the full generated reverse-engineering output set. It systematically challenges generated specs to reduce the risk of incorrect or incomplete specifications.

How do I generate human-validation questions for unresolved spec gaps?

Generating human-validation questions involves applying confidence rules to inspect unit canonical files and optional artifacts. It outputs a structured questions file specifically targeting unresolved critical gaps found during the review phase.

Can I use confidence scoring to reclassify reverse-engineering spec gaps?

Yes, applying confidence scoring reclassifies spec gaps by inspecting unit canonical files and applying confidence rules. It produces a confidence report and a structured questions list for any blocking gaps requiring human validation.

What is the best way to review per-unit spec folders in a reverse-engineering workflow?

Reviewing per-unit spec folders requires reading state.json and config.toml to inspect unit canonical files. It checks business-rule coherence and generates human-validation questions with correct gap handling for the output folder.

Why does my reverse-engineering spec review produce a blocking gap confidence report?

A blocking gap confidence report is produced when applying confidence rules detects unresolved critical inconsistencies in the generated output. It surfaces structured human-validation questions to challenge incorrect or incomplete specifications.