reversa-new

Transform natural-language product ideas into SDD documentation artifacts.

Updated May 16, 2026
One-click install
npx skills add https://github.com/ricardops34/rcg-crm --skill reversa-new
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reversa-new
Source: https://github.com/ricardops34/rcg-crm/tree/main/.agents/skills/reversa-new
Command: npx skills add https://github.com/ricardops34/rcg-crm --skill reversa-new

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It turns a product idea described in natural language into structured SDD specifications, so the forward cycle can start without guesswork.

Core Features & Use Cases

  • Greenfield pipeline orchestration: Coordinates a fixed, multi-agent flow that produces ideation, personas, a PRD, and component-level SDD files.
  • Stateful progress and re-execution control: Reads and updates .reversa/state.json to detect in-progress stages, continue safely, or recreate from selected agents with explicit confirmation.
  • Safety guardrails for artifacts: Writes only to .reversa/ and _reversa_sdd/, and never auto-advances between stages without the user’s CONTINUAR.

Quick Start

Run the command /reversa-new and describe your product idea so the skill can generate ideation.md, personas.md, prd.md, and sdd/*.md inside _reversa_sdd/.

Frequently Asked Questions about reversa-new

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

FAQPage Schema
How do I generate a PRD and SDD from a natural language product idea?

To generate a PRD and SDD from a natural language product idea, you can use a multi-stage agent pipeline that transforms text descriptions into structured ideation, personas, and component-level documentation artifacts.

What is the best way to plan greenfield software development from scratch?

Planning greenfield software development from scratch is best handled by a fixed multi-stage pipeline that sequentially produces ideation, personas, a PRD, and component-level SDD files, ensuring the forward development cycle begins without guesswork.

How do I orchestrate multi-agent pipelines for product specification reconstruction?

Orchestrating multi-agent pipelines for product specification reconstruction involves using stateful stage tracking with a state JSON file, requiring explicit user confirmation to advance between stages and safely recreate artifacts from selected agents.

Can I control agent re-execution and resume progress in a greenfield planning workflow?

Yes, you can control agent re-execution and resume progress in a greenfield planning workflow by reading and updating a state JSON file, allowing you to detect in-progress stages, continue safely, or recreate documentation from selected agents with explicit confirmation.

Do I need YAML frontmatter to activate product specification generation pipelines?

Yes, you need YAML frontmatter to activate the product specification generation pipeline, as the process relies on YAML-frontmatter-driven activation to initialize the deterministic artifact output and stateful tracking for the forward development cycle.

What are the limitations of automated SDD artifact generation pipelines?

Limitations of automated SDD artifact generation pipelines include strict write restrictions to specific directories and an inability to auto-advance between stages, requiring explicit user confirmation before progressing through the fixed multi-agent flow.