reversa

Operates document analysis workflows for managing scanned images and extracting embedded text/code from them.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Reversa solves the problem of legacy systems lacking reliable, up-to-date specifications that AI agents can safely follow to modify or extend production code.

Core Features & Use Cases

  • Central Orchestration: Coordinates a complete reverse-specification workflow by reading persisted state, resuming phases, and sequencing specialized agents one-by-one.
  • Operational Contracts Output: Converts extracted knowledge (business rules, flows, module contracts, architectural decisions) into traceable specs that coding agents can execute with fidelity.
  • Interactive Documentation Control: Guides the user through doc level selection (Essential/Complete/Detailed) and spec organization granularity (module/use-case/endpoint/hybrid/feature/custom), persisting the decision for downstream agents.
  • Safety Guardrails: Enforces a “do not delete or overwrite existing project files” policy by only writing into Reversa-managed folders and state.

Quick Start

Activate the Reversa skill by asking your agent to run "/reversa" to orchestrate a full analysis of your legacy project and generate executable specifications.

Frequently Asked Questions about reversa

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

FAQPage Schema
How do I generate executable AI specs from legacy code?

Legacy code reverse engineering extracts business rules and module contracts into traceable specifications. Reversa orchestrates this by sequentially activating specialized agents to analyze your project and output operational contracts that AI coding agents can safely execute.

What is reverse documentation engineering for legacy modernization?

Reverse documentation engineering for legacy modernization converts undocumented project logic into traceable specifications. It extracts business rules, flows, and architectural decisions, turning them into operational contracts that AI coding agents can execute with fidelity.

How do I checkpoint and resume a legacy code analysis session?

You checkpoint and resume legacy code analysis by persisting session state in a .reversa/state.json file. The orchestration workflow reads this checkpoint to resume phased execution seamlessly, allowing you to stop and continue first-run planning across multiple sessions.

Can I control the documentation level and spec organization for my project?

Yes, you can control documentation level and spec organization granularity through interactive selection. Choose Essential, Complete, or Detailed doc levels, and specify organization by module, use-case, endpoint, hybrid, feature, or custom, persisting decisions for downstream agents.

Does reverse engineering legacy code modify or delete existing project files?

Reverse engineering legacy code does not modify or delete existing project files. Safety guardrails enforce a strict policy where all generated specifications and state checkpoints are written exclusively into managed folders like .reversa/ and _reversa_sdd/.

What is the best way to orchestrate AI agents for legacy code analysis?

The best way to orchestrate AI agents for legacy code analysis is using a central orchestration workflow that sequences specialized agents one-by-one. This approach reads persisted state, resumes phases, and follows step-by-step reference flows to generate executable operational contracts.