infer-logical-architecture-from-code

What problem does it solve?

This Skill turns a codebase scan signal set into a deterministic, technology-agnostic logical architecture proposal, so teams can bootstrap a bounded-context and capability view during brownfield codification without inventing physical implementation details.

Core Features & Use Cases

• Infer bounded contexts from scan evidence: Derives candidate bounded contexts from repository/module structure, idioms, co-change signals, and entry points while keeping boundaries grounded in scan-index data.
• Derive components, responsibilities, and capability coverage: Infers component roles and component-to-capability serving relationships only for the selected capability universe, flagging coverage gaps for review.
• Build a logical architecture artifact with validation guardrails: Produces a complete logical-architecture output (topology claim, data model entities + relationships, capability-level API surface, integration points, ADR log), then enforces an abstraction-layer scrub to prevent leaking physical/implementation details.

Quick Start

Run infer-logical-architecture-from-code during /codify with scan-index.json plus the prior scope, enriched-capabilities, and features proposals to generate logical-architecture.yaml in the codify evidence location for tech-architect validation.