reversa-architect

Synthesize legacy project evidence into C4 diagrams, ERD, and traceability artifacts.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It turns scattered, implicit knowledge inside a legacy codebase into a complete set of operational architectural artifacts that AI coding agents can safely use.

Core Features & Use Cases

  • C4 architecture diagrams (Context → Containers → Components): Produce Mermaid-based C4 outputs that describe the system, its containers, and key internal responsibilities.
  • Complete ERD modeling: Generate an ERD covering main entities, attributes, and relationships with cardinalities and keys.
  • Traceability and impact mapping: Create a Spec Impact Matrix to show which architectural components impact others, improving safe evolution and change planning.
  • Deployment guidance (when available): Optionally produce deployment/infrastructure documentation if deployment-related files/configs exist.

Quick Start

Ask the AI agent to run the reversa-architect skill to synthesize C4 diagrams, ERD, integrations, and the Spec Impact Matrix from the artifacts found in .reversa/state.json and the output folder specified by output_folder.

Frequently Asked Questions about reversa-architect

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

FAQPage Schema
How do I generate C4 architecture diagrams from legacy code?

Generating C4 architecture diagrams from legacy code requires synthesizing reverse-engineered project evidence into Mermaid-based C4 outputs. The process reads project state artifacts to produce deterministic context, container, and component diagrams.

What is a spec impact matrix and when do I need it for reverse engineering?

A spec impact matrix maps which architectural components impact others, improving safe evolution and change planning. You need it during reverse engineering to trace dependencies and assess the blast radius of modifications in legacy systems.

How do I document legacy database schemas into an ERD for AI agents?

Documenting legacy schemas into an ERD involves synthesizing reverse-engineered evidence to generate entity relationship diagrams. This covers main entities, attributes, relationships, cardinalities, and keys to provide operational architectural artifacts for AI coding agents.

Can I use reverse engineering artifacts to produce deployment and infrastructure documentation?

Yes, you can produce deployment and infrastructure documentation from reverse engineering artifacts. This is generated optionally when deployment-related files or configurations exist within the analyzed legacy codebase evidence.

Do I need a state.json file to generate architecture documentation for a legacy codebase?

Yes, you need a state.json file containing output_folder and doc_level configurations. This file provides the required reverse-engineered project evidence and parameters needed to synthesize complete architectural specifications.

What are the limitations of generating architectural specs from legacy code?

Limitations include relying on the accuracy of prior reverse-engineered evidence and the doc_level setting. Deployment documentation is only produced when deployment-related files or configs are explicitly found in the analyzed artifacts.