メニュー
Conduct a rigorous architecture review covering scope, drivers, quality attributes, risks, decisions, and action items. Use when a system, service, or major change needs an ATAM-lite review before commitment.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Build production integrations with Anthropic Claude (Claude 4.5 Sonnet, Claude Opus 4, Claude Haiku 4) via the Messages API — system prompts, multi-turn conversations, tool use (function calling), prompt caching, extended thinking, vision, batched requests, and streaming. Use when wiring Claude into backends, agentic workflows, or evaluating it against other foundation models. Requires ANTHROPIC_API_KEY.
Deploy Cloudflare Workers, Pages, and Workers-with-static-assets safely using Wrangler 4.x with versioned uploads, gradual deployments, preview/production environments, secret bindings, observability, and one-command rollback. Use when shipping edge workloads, API routes, or full-stack apps to Cloudflare. Requires CLOUDFLARE_API_TOKEN with scoped permissions or an authenticated Wrangler session.
Deploy static sites, framework apps (Next.js, Astro, Remix, SvelteKit, Nuxt), Edge Functions, and serverless Functions to Netlify with deploy previews, branch deploys, atomic production promotion, scoped env vars, and instant rollback. Use when releasing on Netlify via CLI, API, or Git-triggered builds. Requires NETLIFY_AUTH_TOKEN (PAT) and NETLIFY_SITE_ID.
Transcribe audio with OpenAI speech-to-text models (gpt-4o-transcribe, gpt-4o-mini-transcribe, whisper-1) using prompt-guided terminology, segment-level timestamps, optional diarization via post-processing, and reliable artifact delivery. Use when converting interviews, calls, podcasts, or meeting recordings into reviewable text. Requires OPENAI_API_KEY.
Deploy Web Services, Private Services, Background Workers, Cron Jobs, Static Sites, PostgreSQL, and Key-Value (Redis-compatible) on Render via Blueprints (`render.yaml`) or the Render API with preview environments, zero-downtime rolling deploys, health checks, env-group secret scoping, and one-click rollback. Use when releasing services, applying schema migrations, or syncing infra-as-code on Render. Requires RENDER_API_KEY (owner- or workspace-scoped).
Operate Sentry end-to-end — error monitoring, performance tracing, session replay, profiling, release health, source-map upload, ownership routing, alert tuning, issue triage, and CI integration via sentry-cli — across web, mobile, and backend stacks. Use when instrumenting an app, wiring releases into deploy pipelines, or running incident-grade triage. Requires SENTRY_AUTH_TOKEN (org-scoped, `project:releases project:write`), project DSN, and (for source maps) `org` + `project` slugs.
| name | architect-review |
| description | Conduct a rigorous architecture review covering scope, drivers, quality attributes, risks, decisions, and action items. Use when a system, service, or major change needs an ATAM-lite review before commitment. |
| risk | unknown |
| source | community |
| date_added | 2026-02-27 |
A structured review surfaces risks before they become incidents. This skill drives a one-to-three day review producing a defensible record of drivers, decisions, sensitivities, trade-offs, and tracked actions aligned to ISO/IEC/IEEE 42010:2022 and ISO/IEC 25010:2023 quality models.
| Concern | Recommended |
|---|---|
| Quality model | ISO/IEC 25010:2023 (functional suitability, reliability, performance efficiency, security, maintainability, portability, compatibility, usability, safety) |
| Architecture description | ISO/IEC/IEEE 42010:2022 viewpoints + arc42 sections |
| Diagrams | C4 model (context, container, component) via Structurizr DSL or Mermaid |
| Decision log | MADR 4.x ADRs in /docs/adr |
| Evaluation method | ATAM-lite (utility tree + scenario walk-through) |
| Risk taxonomy | Risk / Sensitivity Point / Trade-off / Non-risk |
| Tech currency check | ThoughtWorks Tech Radar (latest volume) |
flowchart LR
A[Frame scope and drivers] --> B[Build utility tree]
B --> C[Walk scenarios against architecture]
C --> D[Classify risks, sensitivities, trade-offs]
D --> E[Record ADRs and actions]
E --> F[Publish review report]
ADR snippet (MADR 4.x):
# ADR-0042: Adopt outbox pattern for order events
Status: Accepted (2026-04-30)
Context: Dual-write between Postgres and Kafka risks lost events on partial failure.
Decision: Persist events to an outbox table; relay via Debezium CDC to Kafka.
Consequences: +Atomic with business txn; -Adds CDC operational surface; revisit when EventStoreDB is mature in our org.
| Pitfall | Why it hurts | Mitigation |
|---|---|---|
| Reviewing diagrams without scenarios | Misses runtime risks | Drive review from utility tree scenarios |
| Treating every concern as a Risk | Dilutes signal | Use Risk/Sensitivity/Trade-off/Non-risk taxonomy |
| No measurable responses | Cannot verify later | Require numeric SLO/SLA in every scenario |
| Skipping ADR follow-up | Decisions drift | Every accepted recommendation lands as an ADR |
| Single-stakeholder view | Bias and blind spots | Include product, security, SRE, and on-call |
| Ignoring tech currency | Lock-in to fading stacks | Cross-check choices vs ThoughtWorks Tech Radar |
# Architecture Review: <System> (<YYYY-MM-DD>)
## 1. Scope and Stakeholders
## 2. Business Drivers and Constraints
## 3. Quality Attribute Utility Tree (top scenarios, H/M/L x H/M/L)
## 4. Architecture Snapshot (C4 context + container links)
## 5. Scenario Walk-throughs
## 6. Findings
- Risk Themes (RT-1..n)
- Sensitivity Points
- Trade-offs
- Non-risks
## 7. Decisions (links to ADR-XXXX)
## 8. Action Items (owner, due date, exit criteria)
## 9. Open Questions