reversa-scout

Maps legacy software projects into inventory, dependencies, and surface.json files.

1.5k|383|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/sandeco/reversa --skill reversa-scout-sandeco
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reversa-scout
Source: https://github.com/sandeco/reversa/tree/main/agents/reversa-scout
Command: npx skills add https://github.com/sandeco/reversa --skill reversa-scout-sandeco

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Reversa-scout turns an undocumented legacy codebase into an actionable project inventory so AI agents can analyze safely and effectively.

Core Features & Use Cases

  • Legacy surface mapping: Enumerates the project’s folder structure, excluding noise directories, to establish the starting point for reverse engineering.
  • Technology & framework discovery: Detects languages, main frameworks/libraries, critical dependency versions, and package manager signals from common configuration files.
  • Operational entry-point inventory: Identifies application entry files, environment/config files, CI/CD workflows, Docker artifacts, and key package scripts to understand how the system runs.
  • Database and testing hints (lightweight): Lists database schema/DDL/ORM model candidates and estimates test coverage from test/spec file patterns.
  • Spec organization suggestion: Produces structured surface.json output and recommends an organization granularity (endpoint/module/use-case/hybrid/feature) based on detected signals.

Quick Start

Run the reversa-scout skill at the start of your legacy analysis to generate _reversa_sdd/inventory.md and .reversa/context/surface.json.

Frequently Asked Questions about reversa-scout

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

FAQPage Schema
How do I map the surface of a legacy codebase for AI agents?

Mapping a legacy project surface involves extracting folder structure, detecting frameworks, and identifying entry points to generate a structured surface.json file. This provides AI agents the context needed for safe reverse engineering analysis.

What is the best way to inventory dependencies and frameworks in an undocumented project?

Inventorying dependencies in an undocumented project requires scanning configuration files to detect languages, frameworks, and critical versions. This produces a dependencies.md file outlining package manager signals and tech stack components for AI analysis.

How do I generate a project inventory for reverse documentation?

Generating a project inventory for reverse documentation requires reading .reversa/state.json for settings, excluding noise directories, and outputting an inventory.md file. This establishes the baseline project structure for downstream reverse engineering agents.

Can I detect database schema hints and test coverage signals without running the legacy code?

Yes, you can detect database schema hints and test coverage signals without running code by scanning for DDL/ORM model candidates and test/spec file patterns. This lightweight reconnaissance estimates coverage and identifies database structures safely.

Does reverse engineering legacy code require a specific folder structure to start?

Reverse engineering legacy code does not require a specific folder structure, but it requires excluding known irrelevant directories during enumeration. The process reads .reversa/state.json settings to filter noise and accurately map the existing project surface.