reversa-forward

Detect the active Reversa feature stage and select the next pipeline step.

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

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

It helps you keep a Reversa “forward pipeline” moving by detecting where an active feature currently stands and recommending the correct next skill—without writing or modifying feature artifacts itself.

Core Features & Use Cases

  • Pipeline orchestration: Reads Reversa state and decides the appropriate next step in the sequence from requirements through coding.
  • Stage-aware routing: Detects the feature’s physical stage by checking for the presence and contents of requirements.md, roadmap.md, and actions.md.
  • Safe interaction model: Never auto-executes the next skill; it always asks you to type CONTINUAR, reducing the risk of unintended changes.
  • Greenfield/legado handling: Detects whether reverse-extraction artifacts exist and prepares the expected folders for downstream skills in greenfield mode.

Quick Start

Type /reversa-forward to have the system tell you your current pipeline stage and suggest the exact next command to run for that feature.

Frequently Asked Questions about reversa-forward

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

FAQPage Schema
How do I automate coding pipeline orchestration from requirements through planning?

Coding pipeline orchestration is handled by reading .reversa/state.json and feature artifacts to detect the current stage. It then identifies and recommends the next appropriate agent step without auto-executing, requiring explicit CONTINUAR confirmation to proceed safely.

How does state detection work for feature routing in a coding pipeline?

State detection works by reading .reversa/state.json and checking for the physical presence and contents of requirements.md, roadmap.md, and actions.md. This identifies whether the feature is in requirements, planning, or coding stage for accurate routing.

What's the best way to continue a feature pipeline without modifying existing artifacts?

To continue a feature pipeline without modifying artifacts, use a routing mechanism that reads existing state and recommends the next step. It never writes or modifies feature artifacts itself, only suggesting the next command and waiting for CONTINUAR confirmation.

Can I use agent skills for greenfield projects that lack reverse-extraction artifacts?

Agent skills can be used for greenfield projects lacking reverse-extraction artifacts. The routing system detects their absence and prepares the expected folders for downstream skills automatically, ensuring the pipeline initializes correctly in greenfield mode.

Does pipeline orchestration auto-execute the next agent step when a stage is detected?

Pipeline orchestration does not auto-execute the next agent step. It applies a safe interaction model by always asking you to type CONTINUAR before proceeding, reducing the risk of unintended changes to your feature artifacts.

When do I need to manually organize specs before routing to the next pipeline stage?

You need to organize specs when the routing skill optionally prompts for spec organization during stage detection. This occurs after reading state files and before confirming the next step, ensuring feature artifacts are structured correctly for downstream processing.