5 specialist operations agents for platform engineering, incident response, cloud cost optimization, infrastructure architecture, and security operations. Dispatch when you need production operations expertise.
Inherits all available tools
Additional assets for this skill
This skill inherits all available tools. When active, it can use any tool Claude has access to.
name: using-ops-team description: | 5 specialist operations agents for platform engineering, incident response, cloud cost optimization, infrastructure architecture, and security operations. Dispatch when you need production operations expertise.
trigger: |
skip_when: |
The ring-ops-team plugin provides 5 specialized operations agents. Use them via Task tool with subagent_type:.
See CLAUDE.md and using-ring for canonical workflow requirements and ORCHESTRATOR principle. This skill introduces ops-team-specific agents.
Remember: Follow the ORCHESTRATOR principle from using-ring. Dispatch agents to handle complexity; don't operate tools directly.
CRITICAL: Understand when to use each plugin:
| Domain | Plugin | Agents |
|---|---|---|
| Development Infrastructure | ring-dev-team | devops-engineer (Docker, IaC, CI/CD) |
| Production Operations | ring-ops-team | platform-engineer, incident-responder, etc. |
| Scenario | Use |
|---|---|
| "Set up Dockerfile and docker-compose" | devops-engineer |
| "Configure service mesh for production" | platform-engineer |
| "Create Terraform modules" | devops-engineer |
| "Design multi-region architecture" | infrastructure-architect |
| "Handle production outage" | incident-responder |
| "Optimize cloud costs" | cloud-cost-optimizer |
ALWAYS pause and report blocker for:
| Decision Type | Examples | Action |
|---|---|---|
| Production Changes | Infrastructure modifications | STOP. Change management required. Ask user. |
| Security Incidents | Potential breach | STOP. Security team lead + legal. |
| Cost Commitments | Reserved instance purchases | STOP. Finance approval required. |
| Architecture Decisions | Region selection, DR strategy | STOP. Strategic decision. Ask user. |
You CANNOT make production-impacting decisions autonomously. STOP and ask.
| Misconception | Reality |
|---|---|
| "I can handle this myself" | ORCHESTRATOR principle: dispatch specialists, don't implement directly. This is NON-NEGOTIABLE. |
| "Ops tasks are simple" | Operations has production impact. Specialist oversight is MANDATORY. |
| "Same as DevOps" | dev-team DevOps handles development infrastructure. ops-team handles production operations. |
| "Cost analysis is just math" | Cost optimization requires business context and risk assessment. DISPATCH specialist. |
| "Security is handled by dev reviewers" | Security-reviewer handles code. security-operations handles infrastructure security. BOTH needed. |
Self-sufficiency bias check: If you're tempted to handle operations directly, ask:
If ANY answer is yes -> You MUST DISPATCH the specialist. This is NON-NEGOTIABLE.
If you catch yourself thinking ANY of these, STOP:
| Rationalization | Why It's WRONG | Required Action |
|---|---|---|
| "This is a small operations task" | Small tasks can cause big outages | DISPATCH specialist |
| "I already know how to do this" | Your knowledge != production context | DISPATCH specialist |
| "Just checking logs/metrics" | Log analysis requires domain expertise | DISPATCH specialist |
| "Cost report is straightforward" | Cost optimization needs risk assessment | DISPATCH specialist |
| "Security scan results are clear" | Findings need prioritization and context | DISPATCH specialist |
| "Incident seems minor" | Minor incidents can escalate. Proper triage required. | DISPATCH incident-responder |
These requirements are NON-NEGOTIABLE:
| Requirement | Why It Cannot Be Waived |
|---|---|
| Dispatch to specialist | Specialists have production context |
| Incident documentation | Memory fades, audit trails required |
| Change management | Production changes need oversight |
| Security escalation | Security incidents have legal implications |
| Cost approval chain | Financial commitments need authorization |
User cannot override these. Time pressure cannot override these. "Small task" cannot override these.
When facing pressure to bypass specialist dispatch:
| User Says | Your Response |
|---|---|
| "Production is down, no time for specialist" | "I understand the urgency. Specialist dispatch ensures proper incident response. Dispatching incident-responder with URGENT context now." |
| "Just restart the service quickly" | "Production restarts require change management. Dispatching incident-responder to assess proper remediation." |
| "Cost analysis can wait" | "Cost optimization opportunities have time-value. Dispatching cloud-cost-optimizer for data-driven analysis." |
| "Security finding is false positive" | "All security findings require verified documentation. Dispatching security-operations to properly assess." |
| "I know the architecture, skip review" | "Architecture decisions have long-term impact. Dispatching infrastructure-architect to validate." |
Critical Reminder:
| Agent | Specializations | Use When |
|---|---|---|
platform-engineer | Service mesh, API gateways, developer platforms, self-service infrastructure | Service mesh config, API gateway setup, platform abstractions, developer portals |
incident-responder | Incident management, RCA, post-mortems, blameless culture | Production incidents, outages, incident coordination, root cause analysis |
cloud-cost-optimizer | Cost analysis, RI management, FinOps, tagging | Cost reviews, optimization recommendations, reserved instance planning |
infrastructure-architect | Multi-region, DR, capacity planning, migrations | Architecture design, DR strategy, capacity planning, infrastructure lifecycle |
security-operations | Security audits, compliance, vulnerability management | Security assessments, compliance validation, vulnerability remediation |
Dispatch template:
Task tool:
subagent_type: "{agent-name}"
model: "opus"
prompt: "{Your specific request with context}"
Note: All ops-team agents require model: "opus" for production-grade analysis.
Teams complement each other: Operations handles production, Development handles code, Reviewers handle quality.
If you need multiple specialists (e.g., incident + security), dispatch in parallel:
CORRECT:
Task #1: incident-responder
Task #2: security-operations
(Both run in parallel)
WRONG:
Task #1: incident-responder
(Wait for response)
Task #2: security-operations
(Sequential = 2x slower)
Production incidents DO NOT bypass specialist dispatch:
| Scenario | Wrong Approach | Correct Approach |
|---|---|---|
| Production down | "Fix directly, document later" | Dispatch incident-responder with URGENT flag |
| Security alert | "I'll check the logs" | Dispatch security-operations for proper assessment |
| Cost anomaly | "Probably normal spike" | Dispatch cloud-cost-optimizer to investigate |
Emergency Dispatch Template:
Task tool:
subagent_type: "incident-responder"
model: "opus"
prompt: "URGENT PRODUCTION INCIDENT: [brief context]. [Your specific request]"
Agents: See "5 Operations Specialists" table above.
Skills:
using-ops-team (this) - Plugin introductionops-incident-response - Incident management workflowops-capacity-planning - Capacity planning processops-cost-optimization - Cost optimization workflowops-disaster-recovery - DR planning and testingops-security-audit - Security audit workflowops-platform-onboarding - Service onboarding to platformops-migration-planning - Migration planning processCommands:
/incident - Production incident management/capacity-review - Infrastructure capacity review/cost-analysis - Cloud cost analysis/security-audit - Security audit workflowDispatch based on your need: