Menu
Run a full architecture review — strengths, findings, checklist, ADR recommendations
Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.
Baseado na classificação ocupacional SOC
| name | review |
| description | Run a full architecture review — strengths, findings, checklist, ADR recommendations |
User runs /review followed by a system description, diagram, or design doc.
Run a structured architecture review against the AI Architect's standards. Always complete ALL sections, even if some are "N/A — not yet defined" (that itself is a finding).
System: [name] Date: [today] Reviewer: AI Architect (Claude) Risk Level: [GREEN / AMBER / RED]
2–3 sentences. Lead with the most critical finding.
What's well-designed? Be specific.
| # | Area | Severity | Finding | Recommendation |
|---|---|---|---|---|
| 1 | HIGH/MED/LOW |
Severity guide:
| Check | Status | Notes |
|---|---|---|
| Single points of failure identified | ✅ / ⚠️ / ❌ | |
| Latency budget defined per component | ✅ / ⚠️ / ❌ | |
| PII exposure points mapped | ✅ / ⚠️ / ❌ | |
| Model/prompt versioning strategy | ✅ / ⚠️ / ❌ | |
| Eval & drift detection plan | ✅ / ⚠️ / ❌ | |
| Fallback / circuit breaker patterns | ✅ / ⚠️ / ❌ | |
| Cost model estimated | ✅ / ⚠️ / ❌ | |
| Governance / audit trail | ✅ / ⚠️ / ❌ |
List any decisions surfaced that should be documented as ADRs.
Numbered action list with suggested owners.
Audit a Google ADK tool file against the five known gotchas (DI params, thin wrappers, _j helper, async-wrap blocking calls, httpx context manager) before they bite in production
After creating an ADR, run the cross-referencing checklist — update README Artifacts table, add Related Decisions section to source proposal, cross-link sibling ADRs to each other
Mid-task durability — dump current state and a next-step file before context fills, so the next session resumes without re-deriving where you were
Generate and verify a pre-demo checklist for an ADK/agent project — env vars, GCP auth, embedding swap, ngrok/Slack wiring, dry-run, tests, real-vs-synthetic data
Scan a project's docs (README, CLAUDE.md, ADRs, runbooks) for inconsistent project naming, scope, or framing — catches the "README says retail, CLAUDE.md still says Kroger" class of bug
Refresh a project's status — derive current state from code/tests/git, regenerate the project's status memory entry and (if requested) a status section in the README
