Menu
On-call rotation design, escalation tiers, runbook standards and burnout prevention
Installer avec Codex ou Claude Copiez ce prompt, collez-le dans Codex, Claude ou un autre assistant, puis laissez-le vérifier la page du skill et l'installer pour vous.
Basé sur la classification professionnelle SOC
**Architecture Decision Record (ADR) — Philosophy-First**: Creates well-structured ADRs using a 3-phase process: (1) Define the architectural philosophy and identity of the system, (2) Explore options through the lens of that philosophy, (3) Adversarial review by a devil's advocate agent. Use this skill whenever the user wants to document a technical decision, create an ADR, record why a technology or approach was chosen, compare architectural alternatives, or mentions 'ADR', 'architecture decision', 'decision record', 'technical decision', 'why did we choose', or 'document this decision'. Also trigger when the user is evaluating trade-offs between technical approaches and wants to formalize the reasoning.
**AI/ML Engineering Review**: Reviews AI/ML systems for production readiness — model serving, MLOps pipelines, LLM integration patterns, prompt engineering, evaluation frameworks, and responsible AI. Covers model deployment, feature stores, experiment tracking, monitoring/drift detection, and AI safety. Use when the user mentions ML, AI, machine learning, model, LLM, GPT, Claude, embeddings, RAG, fine-tuning, MLOps, model serving, feature store, or any AI/ML infrastructure.
**API Documentation & Design (OpenAPI/Swagger)**: Helps write and review API documentation, generate OpenAPI/Swagger specs, design RESTful APIs, and document GraphQL schemas. Use whenever the user mentions 'API docs', 'Swagger', 'OpenAPI', 'API specification', 'API design', 'REST API', 'endpoint documentation', 'API contract', 'API versioning', 'GraphQL schema', 'API reference', or asks to document their API, generate a Swagger spec, design API endpoints, or review API contracts for consistency.
API Gateway design patterns including rate limiting, authentication, versioning, BFF and circuit breaking
ARB process design, RFC governance, decision log and technical review workflows
**Backend Code Review (Node.js, Java, Microservices)**: Expert review of backend code focusing on Node.js, Java, Clean Architecture, SOLID principles, microservices patterns, SQL/database design, messaging (Kafka, SQS, SNS), and payment flows. Use whenever the user wants a review of backend code, API design, service architecture, database queries, or mentions Node, Java, Spring, NestJS, Express, microservices, REST API, gRPC, or asks to review server-side code. Also trigger for database schema reviews, query optimization, and message queue patterns.
| name | on-call-design |
| description | On-call rotation design, escalation tiers, runbook standards and burnout prevention |
| category | operational |
| preferred-model | sonnet |
| min-confidence | 0.8 |
| depends-on | ["observability","incident-postmortem"] |
| estimated-tokens | 4000 |
| triggers | {} |
| tags | ["on-call","escalation","runbook","pagerduty"] |
Tier 1 (First Responder):
Who: Engineer on rotation (weekly cycle)
SLA: Acknowledge in 5 min, respond in 15 min
Scope: All production alerts for their service(s)
Tier 2 (Specialist):
Who: Senior engineer / tech lead
SLA: Respond in 30 min
Scope: Escalated issues, cross-service problems
Tier 3 (Leadership):
Who: Engineering manager / CTO
SLA: Respond in 1 hour
Scope: Major incidents, customer-facing outages, data breaches
| Pattern | Team Size | Pros | Cons |
|---|---|---|---|
| Weekly rotation | 4+ engineers | Simple, predictable | Long shifts |
| Follow-the-sun | 8+ (multi-TZ) | No night pages | Coordination overhead |
| Primary/Secondary | 6+ | Backup coverage | Requires 2 people always |
| Business hours only | 3+ | No night pages | Gaps in coverage |
| Hybrid | 5+ | Balanced | Complex scheduling |
Week 1: Alice (primary), Bob (secondary)
Week 2: Bob (primary), Carol (secondary)
Week 3: Carol (primary), Dave (secondary)
Week 4: Dave (primary), Alice (secondary)
Rules:
- Primary: first responder, expected to resolve or escalate
- Secondary: backup if primary doesn't acknowledge in 10 min
- Handoff: Monday 10am (overlap with both people)
- Minimum 4 weeks between primary shifts
- Never on-call during PTO
# PagerDuty/OpsGenie-style escalation
escalation_policy:
name: "Backend API"
repeat_count: 2 # Retry full chain 2x before giving up
steps:
- timeout_minutes: 5
targets:
- type: on_call_primary
- timeout_minutes: 10
targets:
- type: on_call_secondary
- timeout_minutes: 15
targets:
- type: team_lead
- notification: slack_channel_critical
- timeout_minutes: 30
targets:
- type: engineering_manager
- notification: phone_call
Every service MUST have: