reversa-sync

Generates post-coding addenda that keep legacy extraction artifacts synchronized with delivered features.

1.5k|405|Updated Apr 26, 2026
One-click install
npx skills add https://github.com/sandeco/reversa --skill reversa-sync
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: reversa-sync
Source: https://github.com/sandeco/reversa/tree/main/agents/reversa-sync
Command: npx skills add https://github.com/sandeco/reversa --skill reversa-sync

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

After a forward-cycle feature is coded, the original extraction artifacts (architecture.md, domain.md) become stale because the code has already changed. This Skill closes that gap by distilling each delivered feature into an addendum file, keeping the extraction readable and accurate until the next full re-extraction.

Core Features & Use Cases

  • Addendum Generation: Creates a dated addendum per delivered feature in _reversa_sdd/addenda/, summarizing impacts per extraction artifact with typed deltas (regra-alterada, componente-novo, delta-de-contrato-externo, etc.).
  • Scenario Detection: Distinguishes legacy scenarios (anchored to architecture.md and domain.md) from greenfield scenarios (anchored to prd.md and sdd/ specs) and adapts the impact table accordingly.
  • Read-Only Safety: Never modifies original extraction artifacts; writes only inside the addenda folder, appending update sections on re-execution instead of overwriting.
  • Use Case: After running /reversa-coding on a feature that changed a business rule, run this Skill to produce an addendum pointing readers to the new rule behavior until the next /reversa re-extraction supersedes it.

Quick Start

Run the reversa-sync skill after /reversa-coding completes to generate the convergence addendum for the active feature.

Frequently Asked Questions about reversa-sync

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

FAQPage Schema
How do I keep reverse-engineered specs in sync after coding a feature?

Run reversa-sync after /reversa-coding completes. It reads the feature's legacy-impact.md and generates a dated addendum in _reversa_sdd/addenda/ describing how each extraction artifact should now be read, valid until the next re-extraction.

What is an addendum in the Reversa forward cycle?

An addendum is a per-feature Markdown file in _reversa_sdd/addenda/ that records delivery impacts on extraction artifacts using typed deltas like regra-alterada or componente-novo. It stays valid until a new /reversa re-extraction marks it as superseded.

Does reversa-sync modify the original architecture or domain documents?

No. The skill writes only inside _reversa_sdd/addenda/ and treats architecture.md, domain.md, prd.md, and sdd specs as strictly read-only. On re-execution it appends an update section rather than overwriting prior addendum content.

What happens if I run reversa-sync before /reversa-coding?

The skill aborts with a message stating the active feature has not passed through /reversa-coding and there is no delivery to converge. It requires feature-dir/legacy-impact.md to exist before generating any addendum.

Can reversa-sync handle greenfield projects without legacy code?

Yes. It detects greenfield scenarios via the legacy-impact.md header or the presence of prd.md and sdd specs, then points impacts to those artifacts with the componente-novo type instead of legacy anchors.