7 specialist developer agents for backend (Go/TypeScript), DevOps, frontend, design, QA, and SRE. Dispatch when you need deep technology expertise.
/plugin marketplace add lerianstudio/ring/plugin install ring-dev-team@ringThis skill inherits all available tools. When active, it can use any tool Claude has access to.
The ring-dev-team plugin provides 7 specialized developer 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 dev-team-specific agents.
Remember: Follow the ORCHESTRATOR principle from using-ring. Dispatch agents to handle complexity; don't operate tools directly.
<block_condition>
If any condition applies, STOP and ask user.
always pause and report blocker for:
| Decision Type | Examples | Action |
|---|---|---|
| Technology Stack | Go vs TypeScript for new service | STOP. Check existing patterns. Ask user. |
| Architecture | Monolith vs microservices | STOP. This is a business decision. Ask user. |
| Infrastructure | Cloud provider choice | STOP. Check existing infrastructure. Ask user. |
| Testing Strategy | Unit vs E2E vs both | STOP. Check QA requirements. Ask user. |
You CANNOT make technology decisions autonomously. STOP and ask.
See shared-patterns/shared-anti-rationalization.md for universal anti-rationalizations (including Specialist Dispatch section).
Self-sufficiency bias check: If you're tempted to implement directly, ask:
If any answer is yes → You MUST DISPATCH the specialist. This is NON-NEGOTIABLE.
See shared-patterns/shared-anti-rationalization.md for universal anti-rationalizations (including Specialist Dispatch section and Universal section).
<cannot_skip>
These requirements are NON-NEGOTIABLE:
| Requirement | Why It Cannot Be Waived |
|---|---|
| Dispatch to specialist | Specialists have standards loading, you don't |
| 6-gate development cycle | Gates prevent quality regressions |
| Parallel reviewer dispatch | Sequential review = 3x slower, same cost |
| TDD in Gate 0 | Test-first ensures testability |
| User approval in Gate 5 | Only users can approve completion |
User cannot override these. Time pressure cannot override these. "Simple task" cannot override these.
See shared-patterns/shared-pressure-resistance.md for universal pressure scenarios (including Combined Pressure Scenarios and Emergency Response).
Critical Reminder:
See shared-patterns/shared-pressure-resistance.md → Emergency Response section for the complete protocol.
Emergency Dispatch Template:
Task tool:
subagent_type: "ring-dev-team:backend-engineer-golang"
model: "opus"
prompt: "URGENT PRODUCTION INCIDENT: [brief context]. [Your specific request]"
IMPORTANT: Specialist dispatch takes 5-10 minutes, not hours. This is NON-NEGOTIABLE even under CEO pressure.
See shared-patterns/shared-pressure-resistance.md → Combined Pressure Scenarios section.
<dispatch_required agent="{specialist}" model="opus"> Use Task tool to dispatch appropriate specialist based on technology need. </dispatch_required>
| Agent | Specializations | Use When |
|---|---|---|
backend-engineer-golang | Go microservices, PostgreSQL/MongoDB, Kafka/RabbitMQ, OAuth2/JWT, gRPC, concurrency | Go services, DB optimization, auth/authz, concurrency issues |
backend-engineer-typescript | TypeScript/Node.js, Express/Fastify/NestJS, Prisma/TypeORM, async patterns, Jest/Vitest | TS backends, JS→TS migration, NestJS design, full-stack TS |
devops-engineer | Docker/Compose, Terraform/Helm, cloud infra, secrets management | Containerization, local dev setup, IaC provisioning, Helm charts |
frontend-bff-engineer-typescript | Next.js API Routes BFF, Clean/Hexagonal Architecture, DDD patterns, Inversify DI, repository pattern | BFF layer, Clean Architecture, DDD domains, API orchestration |
frontend-designer | Bold typography, color systems, animations, unexpected layouts, textures/gradients | Landing pages, portfolios, distinctive dashboards, design systems |
qa-analyst | Test strategy, Cypress/Playwright E2E, coverage analysis, API testing, performance | Test planning, E2E suites, coverage gaps, quality gates |
sre | Structured logging, tracing, health checks, observability | Logging validation, tracing setup, health endpoint verification |
Dispatch template:
Task tool:
subagent_type: "{agent-name}"
model: "opus"
prompt: "{Your specific request with context}"
Note: frontend-designer = visual aesthetics. frontend-bff-engineer-typescript = business logic/architecture.
Both can be used together: Get developer specialist guidance during design, then run general reviewers before merge.
If you need multiple specialists (e.g., backend engineer + DevOps engineer), dispatch in parallel (single message, multiple Task calls):
✅ CORRECT:
Task #1: backend-engineer-golang
Task #2: devops-engineer
(Both run in parallel)
❌ WRONG:
Task #1: backend-engineer-golang
(Wait for response)
Task #2: devops-engineer
(Sequential = 2x slower)
Remember:
"I need a Go service. Let me dispatch
backend-engineer-golangto design it."
"I'll manually read Go best practices and design the service myself."
Agents: See "7 Developer Specialists" table above.
Skills: using-dev-team (this), dev-cycle (6-gate workflow), dev-refactor (codebase analysis)
Commands: /dev-cycle (execute tasks), /dev-refactor (analyze codebase)
Note: Missing agents? Check .claude-plugin/marketplace.json for ring-dev-team plugin.
All workflows converge to the 6-gate development cycle:
| Workflow | Entry Point | Output | Then |
|---|---|---|---|
| New Feature | /pre-dev-feature "description" | docs/pre-dev/{feature}/tasks.md | → /dev-cycle tasks.md |
| Direct Tasks | /dev-cycle tasks.md | — | Execute 6 gates directly |
| Refactoring | /dev-refactor | docs/refactor/{timestamp}/tasks.md | → /dev-cycle tasks.md |
6-Gate Development Cycle:
| Gate | Focus | Agent(s) |
|---|---|---|
| 0: Implementation | TDD: RED→GREEN→REFACTOR | backend-engineer-*, frontend-bff-engineer-typescript |
| 1: DevOps | Dockerfile, docker-compose, .env | devops-engineer |
| 2: SRE | Health checks, logging, tracing | sre |
| 3: Testing | Unit tests, coverage ≥85% | qa-analyst |
| 4: Review | 3 reviewers IN PARALLEL | code-reviewer, business-logic, security |
| 5: Validation | User approval: APPROVED/REJECTED | User decision |
Key Principle: All development follows the same 6-gate process.
Dispatch based on your need:
This skill should be used when the user asks to "create a slash command", "add a command", "write a custom command", "define command arguments", "use command frontmatter", "organize commands", "create command with file references", "interactive command", "use AskUserQuestion in command", or needs guidance on slash command structure, YAML frontmatter fields, dynamic arguments, bash execution in commands, user interaction patterns, or command development best practices for Claude Code.
This skill should be used when the user asks to "create an agent", "add an agent", "write a subagent", "agent frontmatter", "when to use description", "agent examples", "agent tools", "agent colors", "autonomous agent", or needs guidance on agent structure, system prompts, triggering conditions, or agent development best practices for Claude Code plugins.
This skill should be used when the user asks to "create a hook", "add a PreToolUse/PostToolUse/Stop hook", "validate tool use", "implement prompt-based hooks", "use ${CLAUDE_PLUGIN_ROOT}", "set up event-driven automation", "block dangerous commands", or mentions hook events (PreToolUse, PostToolUse, Stop, SubagentStop, SessionStart, SessionEnd, UserPromptSubmit, PreCompact, Notification). Provides comprehensive guidance for creating and implementing Claude Code plugin hooks with focus on advanced prompt-based hooks API.