← ClaudeAtlas

thalarch-architecturelisted

Evidence-driven software architecture design and review. Use for module/service boundaries, dependency direction, monolith vs distributed decomposition, system design, scalability, platform/data decisions, ADRs, architecture refactors, or cross-cutting changes where tradeoffs and quality attributes matter more than local code style.
LUC4N3X/antigravity-thalarch · ★ 2 · Code & Development · score 65
Install: claude install-skill LUC4N3X/antigravity-thalarch
# Thalarch Architecture Architecture is a set of costly-to-change decisions and explicit tradeoffs, not a catalog of patterns to apply by fashion. ## 1. Start from current reality Before proposing architecture, establish from repository/runtime evidence: - existing modules/services and ownership boundaries; - entry points and dependency directions; - data stores and consistency boundaries; - deployment units and runtime topology when known; - public/internal APIs and event/message contracts; - build/test/release constraints; - operational observability and failure domains; - important historical constraints from docs/ADRs/Git when available. For an existing project, prefer evolutionary changes over greenfield redesign unless the user explicitly requests replacement architecture. ## 2. Quality-attribute contract Make the forces explicit before choosing a pattern: - correctness/consistency; - latency/throughput; - availability/resilience; - security/privacy; - scalability; - deployability; - operability/observability; - testability; - maintainability/change frequency; - data ownership/migration cost; - team/organizational constraints when actually known; - cost and platform constraints. Do not rank an architecture without saying which attributes it optimizes and what it sacrifices. ## 3. Alternatives, not foregone conclusions For material decisions compare at least the plausible alternatives, including the simplest option. For each alternative record: - benefits un