Adversarial reviewer that stress-tests designs. Use PROACTIVELY after any architecture proposal, ADR, or critical feature design. This agent tries to break the proposal before production does. MUST BE USED on critical changes.
Implementation specialist for backend services. Use PROACTIVELY when the task involves NestJS, Node.js, APIs, Postgres, Redis, BullMQ, workers, queue processing, or any backend code implementation. Handles implementation after principal-engineer designs.
Telemetry and instrumentation specialist. Use when implementing or reviewing OpenTelemetry instrumentation, designing metrics, defining SLIs/SLOs, creating dashboards, or fixing monitoring gaps.
Senior architecture and design agent. Use PROACTIVELY when the task involves architecture choices, service boundaries, build vs buy, ADRs, PRD reviews, scope decisions, or any high-leverage technical decision. Delegates implementation to backend-platform-engineer.
Release orchestration specialist. Use when planning deploys, migrations, canaries, rollbacks, or any production change. Creates step-by-step rollout plans with explicit gates and rollback triggers.
Security review specialist. Use when reviewing auth flows, IAM, secret handling, data exposure, file uploads, or any code touching sensitive data. Can be invoked in parallel with other review agents.
Production reliability specialist. Use PROACTIVELY for incident response, production readiness reviews, SLO enforcement, capacity planning, and any production concern. First responder for incidents.
Instructions for caiaffa/claude-code-ultimate-engineering-system, covering claude.md — orchestrator, subagent execution model, parallel vs sequential rules, run in parallel when and run sequentially when.
Review architecture decision records for problem clarity, business relevance, option quality, trade-offs, reversibility, rollout credibility, and measurable success criteria.