archReview

Review architecture plans for technical soundness, diagrams, and edge case analysis.

1|2|Updated Mar 19, 2026
One-click install
npx skills add https://github.com/joi-fairshare/agentic-workflow --skill archreview
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: archReview
Source: https://github.com/joi-fairshare/agentic-workflow/tree/main/skills/archReview
Command: npx skills add https://github.com/joi-fairshare/agentic-workflow --skill archreview

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

This skill catches architectural flaws, missing diagrams, and unexamined failure modes before plans are implemented, replacing guesswork with a structured verdict.

Core Features & Use Cases

  • Diagram Mandates: Automatically require component, data flow, and sequence mermaid diagrams to make boundaries and interactions explicit.
  • Edge Case Coverage: Enumerates dependency outages, adversarial inputs, load stress, and state leakage risks for each component boundary.
  • Decision Audit Trail: Saves structured prism-mcp context, ledger entries, and risk tables so engineering teams have an evidence-backed verdict.

Quick Start

Run /archReview against the latest plan directory to audit the architecture immediately.

Frequently Asked Questions about archReview

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

FAQPage Schema
How do I review an architecture plan for missing diagrams and edge cases?

Architecture plan review evaluates technical soundness by mandating component, data flow, and sequence diagrams while enumerating dependency outages and adversarial inputs to catch flaws before implementation.

What is the best way to audit technical soundness in a repo architecture plan?

Auditing technical soundness applies structured verdicts against repository plans or directories, checking that risk tables and edge case matrices satisfy predefined architecture review checkpoints.

Can I generate mermaid diagrams for my data flow and sequence boundaries during a review?

Yes, architecture review mandates mermaid diagrams for component, data flow, and sequence interactions, making system boundaries and component interactions explicit within the generated verdict.

Does architecture review work for identifying failure modes like state leakage and load stress?

Architecture review identifies failure modes by enumerating dependency outages, adversarial inputs, load stress, and state leakage risks for each component boundary within an edge case matrix.

How do I save an audit trail for architecture decisions and risk tables?

Saving an architecture decision audit trail stores structured context, ledger entries, and risk tables, providing engineering teams with an evidence-backed verdict for future reference.

Why should I not use manual guesswork for architecture risk analysis?

Manual guesswork lacks structured evidence, whereas architecture risk analysis replaces it by applying edge case coverage and diagram mandates to produce an evidence-backed verdict.