reversa-coding

Executes actions.md tasks into code and generates legacy-impact and regression-watch audit artifacts.

Updated Aug 14, 2026
One-click install
npx skills add https://github.com/Ruggery/OCR_Cartorio --skill reversa-coding-ruggery
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: reversa-coding
Source: https://github.com/Ruggery/OCR_Cartorio/tree/main/.agents/skills/reversa-coding
Command: npx skills add https://github.com/Ruggery/OCR_Cartorio --skill reversa-coding-ruggery

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Turning a planned task list into actual code often loses traceability: nobody records which legacy rules were touched, which behaviors must keep holding true, or what failed mid-run. This Skill executes the actions.md plan phase by phase while producing a complete audit trail for future re-extraction cycles. ## Core Features & Use Cases - Phased task execution: Runs unchecked actions from actions.md in order (Preparação, Testes, Núcleo, Integração, Polimento), respecting dependencies and parallel markers, checking off [X] boxes and appending to progress.jsonl. - Legacy impact report: Generates legacy-impact.md mapping every touched file to architecture.md components with impact type and severity, listing preserved and modified domain rules. - Regression watch list: Creates regression-watch.md with stable watch IDs for modified or removed rules, so future re-extractions can verify nothing silently broke. - Dual context anchoring: Works anchored on legacy extraction artifacts (architecture.md + domain.md) or greenfield artifacts (prd.md + SDD specs), aborting cleanly when neither exists. - Use Case: After running /reversa-to-do to produce an actions.md plan for a new feature, invoke this Skill to implement the tasks, then hand off to /reversa-sync for convergence. ## Quick Start Ask the agent to run reversa-coding to execute the pending actions in the current feature directory and generate the impact and regression reports.

Frequently Asked Questions about reversa-coding

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

FAQPage Schema
How do I execute an actions.md implementation plan with an AI agent?▼

Run the reversa-coding skill after generating actions.md with /reversa-to-do. It executes unchecked actions phase by phase, marks each completed item [X], logs progress.jsonl entries, and stops to report if any action fails.

What artifacts does reversa-coding generate after implementation?▼

It generates legacy-impact.md, mapping touched files to architecture components with impact type and severity, and regression-watch.md, listing rules that must keep holding true with stable watch IDs. It also updates actions.md checkboxes and appends to progress.jsonl.

Can reversa-coding run on a greenfield project without legacy code?▼

Yes, it anchors on greenfield artifacts when _reversa_sdd/ contains prd.md and at least one spec in the sdd/ folder. The impact and watch files are still generated, adapted to reference SDD specs instead of extracted legacy rules.

Why does reversa-coding abort before writing any code?▼

It aborts when no context anchor exists in _reversa_sdd/: neither legacy files (architecture.md plus domain.md) nor greenfield files (prd.md plus SDD specs). It also aborts if active-requirements.json or actions.md is missing, pointing to the prerequisite command.

Can I run only part of the actions.md plan?▼

Yes, pass a free-form argument naming a phase or an ID interval such as T001-T005 to restrict execution scope. Without an argument, it runs all remaining unchecked actions in the standard phase order.