Architecture review

Skill AmanKrSahu/zeroday-auditor/skills/architecture-review

An enterprise-grade, multi-specialist engineering audit plugin for Claude Code.

Install

npx -y skills add AmanKrSahu/zeroday-auditor --skill architecture-review

Assembled from the repository path, not quoted from the project. Check it against their README if it does not work.

One thing to look at

0 stars0 stars. Stars are a popularity signal and not a quality one, but at this level it is likely that nobody has read this closely except its author, and you would be relying on your own review.

What its author says it does

Copied from the file, not written here

Use when the user wants a focused architecture/design review of a codebase — 'review the architecture', 'assess the system design', 'is this well structured?', 'evaluate modularity/coupling/layering', 'map the request lifecycle and dependencies', 'will this scale architecturally?'. Produces a single evidence-backed report (architecture-review.md) with a structure map, request/auth flow, dependency analysis, risks, and improvement proposals. For a full multi-dimension audit use zeroday-auditor instead. Analysis only; never modifies source.

SKILL.md

3.3 KB, as published. Nobody here has run it

Architecture Review

Perform a focused architecture and design review as a Principal Software Architect + Senior Technical Lead + Senior API Architect. Output one self-contained report, architecture-review.md, in the repository root. Analysis only — never modify source.

Ground every observation in the actual code. You cannot assess a design you haven't traced — read enough that you could describe the request lifecycle end-to-end yourself, not just from subagent summaries. Where evidence is thin, say "Insufficient evidence to conclude."

Workflow

Understand the system. Map the tree; read the entrypoint, routing/controller/service layout, config, ORM schema, and README/docs. Determine purpose, architecture style, deployment model, request lifecycle, auth flow, state management, and the design patterns/conventions in use.
Dispatch specialist passes (concurrently) via audit-specialist (fallback general-purpose), READ-ONLY, returning evidence with file:line:
- Structure & boundaries — layering, separation of concerns, coupling/cohesion, module responsibilities, dependency direction (and any cycles), abstraction leaks.
- Flows & contracts — request lifecycle end-to-end, auth flow, API surface/versioning, data access patterns, cross-cutting concerns (error handling, config, logging placement).
- Scale & evolvability — statefulness, horizontal-scaling readiness, extension points, and where change is expensive.
Verify the load-bearing claims yourself (e.g. an alleged dependency cycle or a layering violation — open the files and confirm). Reconcile duplicates.
Write architecture-review.md (structure below).

Finding format

Give architectural risks stable IDs ARCH-<C|H|M|L>-NNN with: Title · Severity · Description · Evidence (file:line) · Root Cause · Impact · Affected Modules · Recommendation · Effort · Risk of change. Separate fact from recommendation. Rate by realistic impact in this system.

Report structure

# <Project> — Architecture Review
(scope, method, date)
# Overview        (purpose, category, stack, deployment model)
# Architecture    (style + a Mermaid/ASCII diagram)
# Request Lifecycle & Auth Flow   (end-to-end)
# Major Modules & Responsibilities
# Dependency Relationships   (and any cycles/leaks)
# Strengths
# Weaknesses & Architectural Risks   (ARCH-* findings)
# Scalability Observations
# Maintainability Observations
# Suggested Architectural Improvements   (prioritized, with rationale and migration notes)

Finish by telling the user the top strengths, the top architectural risks, and the highest-leverage improvement. Confirm source is untouched (git status shows only architecture-review.md).