speckit-arch-reverse

Reverse-generate 4+1 architecture views from observable repository facts.

Updated Mar 16, 2026
One-click install
npx skills add https://github.com/B0yZ4kr14/OrthoPlus-Enterprise --skill speckit-arch-reverse
Or copy as Structured Prompt for Agent▼
Please help me install this Agent Skill.
Skill: speckit-arch-reverse
Source: https://github.com/B0yZ4kr14/OrthoPlus-Enterprise/tree/main/.specify-backups/20260518-141826/_agents/skills/speckit-arch-reverse
Command: npx skills add https://github.com/B0yZ4kr14/OrthoPlus-Enterprise --skill speckit-arch-reverse

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Speckit-arch-reverse turns an existing (historical) repository into an updated, architecture-level 4+1 set of artifacts by using observable repository facts as evidence, so teams can get a single source of truth for architecture without doing manual reverse engineering.

Core Features & Use Cases

  • Evidence-first reverse architecture: Records repository facts as an explicit evidence layer to support architecture conclusions rather than turning views into audit reports.
  • 4+1 architecture synthesis refresh: Generates/refreshes scenario, logical, process, development, and physical architecture views plus the main architecture SSOT.
  • Gap-aware output: Writes specific missing-evidence gaps into the facts and affected views instead of inventing unknown architecture elements.

Use case: You join a team inheriting a large monorepo and need an architecture SSOT quickly for alignment and onboarding.

Quick Start

Ask the skill to reverse-generate the architecture views from the current repository state by providing any required user input for arguments.

Frequently Asked Questions about speckit-arch-reverse

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

FAQPage Schema
How do I generate 4+1 architecture views from an existing repository?▼

To generate 4+1 architecture views from an existing repository, this skill captures observable repository facts from file structures and configurations, then synthesizes scenario, logical, process, development, and physical views into a single source of truth.

What is the best way to reverse engineer software architecture documentation for a monorepo?▼

Reverse engineering software architecture documentation is best handled by capturing repo evidence first, then synthesizing the 4+1 views to produce an architecture SSOT, avoiding manual reconstruction and ensuring alignment for onboarding teams.

Can I update outdated architecture documentation using repository evidence?▼

Yes, you can update outdated architecture documentation by inspecting file structures, docs, and operational manifests as evidence, which refreshes the 4+1 views and explicitly reports missing evidence gaps instead of inventing unknown elements.

How do I run the setup script to locate output paths for architecture synthesis?▼

To locate output paths for architecture synthesis, you must run the repo-structure setup script first, which ensures the seven specified .specify/memory/* files are written correctly for the 4+1 views generation process.

Does reverse architecture generation work when configuration or operational manifests are missing?▼

Yes, reverse architecture generation works with missing configurations by enforcing evidence grounding and writing specific missing-evidence gaps into the facts and affected views, rather than fabricating unknown architecture elements.

Why does reverse architecture synthesis require an evidence layer before generating views?▼

Reverse architecture synthesis requires an evidence layer to support architecture conclusions with observable repository facts, preventing the views from becoming audit reports and ensuring the 4+1 architecture SSOT remains grounded.