orchestrator-design
Design O-Agent systems for multi-agent fleet management. Use when building orchestrator agents, designing multi-agent architectures, or creating unified interfaces for agent fleet control.
What this skill does
# Orchestrator Design Skill Design O-Agent (Orchestrator Agent) systems for managing fleets of specialized agents. ## Purpose Guide the architectural design of orchestrator systems that create, command, monitor, and delete specialized agents through a single unified interface. ## When to Use - Designing multi-agent systems - Building fleet management architecture - Creating scalable agent workflows - Implementing the Single Interface Pattern ## Prerequisites - Understanding of the Three Pillars (@three-pillars-orchestration.md) - Familiarity with agent lifecycle patterns (@agent-lifecycle-crud.md) - Access to Claude Agent SDK documentation ## SDK Requirement > **Implementation Note**: Orchestrator patterns require Claude Agent SDK with custom MCP tools. Claude Code subagents cannot spawn other subagents. ## Design Process ### Step 1: Define Orchestration Scope Answer these questions: - What workflows will be orchestrated? - What agent types are needed? - What is the expected scale? - What observability is required? **Output**: Scope document (requirements and constraints) ### Step 2: Design Agent Templates For each agent type needed: | Template | Purpose | Model | Tools | | --- | --- | --- | --- | | scout-fast | Quick reconnaissance | Haiku | Read, Glob, Grep | | builder | Code implementation | Sonnet | Read, Write, Edit, Bash | | reviewer | Code review | Sonnet | Read, Grep, Glob, Bash | | planner | Task planning | Sonnet | Read, Glob, Grep | **Template Structure**: ```yaml --- name: template-name description: What this agent does tools: [tool1, tool2] model: sonnet|haiku --- # System Prompt [Agent-specific instructions] ``` ### Step 3: Design Orchestrator System Prompt The orchestrator needs a specific identity: ```markdown # Orchestrator Agent ## Purpose Manage and coordinate specialized agents to accomplish complex tasks. You do NOT perform work directly - you orchestrate other agents. ## Capabilities - Create specialized agents from templates - Command agents with detailed prompts - Monitor agent progress - Aggregate and report results - Delete agents when work is complete ## Workflow Pattern 1. Analyze task requirements 2. Create appropriate agents 3. Command agents with detailed instructions 4. Monitor progress 5. Aggregate results 6. Report to user 7. Delete agents ## Context Protection - Keep your context focused on orchestration - Delegate detailed work to specialized agents - Do not read files directly - Do not write code ``` ### Step 4: Define Management Tools Design the MCP tools for agent management: | Tool | Purpose | Parameters | | --- | --- | --- | | `create_agent` | Spin up new agent | template, name | | `command_agent` | Send prompt to agent | agent_id, prompt | | `check_agent_status` | Get agent progress | agent_id | | `list_agents` | View all active agents | - | | `delete_agent` | Clean up agent | agent_id | | `read_agent_logs` | View agent responses | agent_id | ### Step 5: Design Observability Layer Essential metrics to track: | Metric | Purpose | | --- | --- | | Agent status | Know what's running | | Context usage | Monitor token consumption | | Costs | Track spend per agent | | Tool calls | See what agents are doing | | Results | Verify outputs | | Time | Measure execution duration | **Observability Components**: 1. Agent Cards - Status, model, context, cost per agent 2. Event Stream - Real-time log of all activities 3. Cost Tracking - Per-agent and total costs 4. Result Inspector - consumed/produced assets 5. Log Viewer - Filterable activity history ### Step 6: Design Workflow Phases Standard orchestration workflow: ```text Phase 1: Scout ├── Create scouts (parallel) ├── Command each with specific area ├── Monitor until complete └── Aggregate findings Phase 2: Build ├── Create builder ├── Command with scout reports ├── Monitor implementation └── Aggregate changes Phase 3: Review ├── Create reviewer ├── Command to verify implementation ├── Monitor review └── Generate final report Cleanup: Delete all agents ``` ### Step 7: Plan Deployment Architecture Required components for SDK implementation: | Component | Purpose | | --- | --- | | Claude Agent SDK | Core orchestration | | MCP Servers | Agent management tools | | Database | Agent state persistence | | WebSocket | Real-time updates | | UI/CLI | User interface | ## Output Format When designing an orchestrator system, provide: ```markdown ## Orchestrator System Design **Name:** [system-name] **Purpose:** [1-2 sentences] **Scale:** [expected agent count and concurrency] ### Agent Templates | Template | Purpose | Model | Tools | | --- | --- | --- | --- | | ... | ... | ... | ... | ### Orchestrator Configuration **System Prompt:** [included or file reference] **Management Tools:** [list of MCP tools] **Observability:** [metrics and components] ### Workflow Design [Phase diagram with agent creation/deletion points] ### Architecture [Deployment diagram with components] ### Implementation Notes [SDK considerations, constraints, scaling factors] ``` ## Design Checklist - [ ] Orchestration scope defined - [ ] Agent templates designed - [ ] Orchestrator system prompt written - [ ] Management tools specified - [ ] Observability layer planned - [ ] Workflow phases designed - [ ] Deployment architecture planned ## Anti-Patterns | Avoid | Why | Instead | | --- | --- | --- | | Orchestrator doing work | Context pollution | Delegate everything | | Missing observability | Flying blind | Track all metrics | | Keeping dead agents | Resource waste | Delete when done | | No lifecycle management | Can't scale | CRUD operations | | Generic agents | Unfocused work | Specialized templates | ## Cross-References - @three-pillars-orchestration.md - Framework foundation - @single-interface-pattern.md - O-Agent architecture - @agent-lifecycle-crud.md - Lifecycle management - @multi-agent-context-protection.md - Context boundaries - @results-oriented-engineering.md - Result patterns ## Version History - **v1.0.0** (2025-12-26): Initial release --- ## Last Updated **Date:** 2025-12-26 **Model:** claude-opus-4-5-20251101
Related in Design
contribute
IncludedLocal-only OSS contribution command center. Auto-refreshes the user's in-flight PR and issue state on invoke so conversations start with full context — no need to brief Claude on what's in flight. Helps the user find issues to contribute to on GitHub, builds per-repo dossiers of what each upstream expects (CLA, DCO, branch convention, AI policy, draft-first, review bots, issue templates), runs deterministic gates before any external action so AI-assisted contributions don't reach maintainers as slop. State is markdown-only: candidate files at ~/.contribute-system/candidates/, repo dossiers at ~/.contribute-system/research/, append-only event log at ~/.contribute-system/log.jsonl. No database, no cloud calls. Use when the user asks about their PRs / issues / contributions, wants to find new work to take on, claim an issue, build/refresh a repo's dossier, or draft a Design Issue or PR. Trigger with "/contribute", "what's my PR status", "find a contribution", "claim issue X", "draft a Design Issue for Y", "refresh dossier for Z".
architectural-analysis
IncludedUser-triggered deep architectural analysis of a codebase or scoped subtree across eight modes — information architecture, data flow, integration points, UI surfaces, interaction patterns, data model, control flow, and failure modes. This skill should be used when the user asks to "diagram this codebase," "map the architecture," "show the data flow," "give me an ERD," "trace control flow," "find the integration points," "verify the layout pattern," "audit the UX architecture," or any similar request whose primary deliverable is mermaid diagrams plus cited reports under docs/architecture/. Dispatches haiku/sonnet sub-agents in parallel for per-mode exploration, then verifies every citation mechanically before any node lands in a diagram. Not for one-off prose explanations of code (use code-explanation) or for high-level system design from scratch (use system-design).
mcp
IncludedModel Context Protocol (MCP) server development and tool management. Languages: Python, TypeScript. Capabilities: build MCP servers, integrate external APIs, discover/execute MCP tools, manage multi-server configs, design agent-centric tools. Actions: create, build, integrate, discover, execute, configure MCP servers/tools. Keywords: MCP, Model Context Protocol, MCP server, MCP tool, stdio transport, SSE transport, tool discovery, resource provider, prompt template, external API integration, Gemini CLI MCP, Claude MCP, agent tools, tool execution, server config. Use when: building MCP servers, integrating external APIs as MCP tools, discovering available MCP tools, executing MCP capabilities, configuring multi-server setups, designing tools for AI agents.
react-native-skia
IncludedDesign, build, debug, and optimise high-polish animated graphics in React Native or Expo using @shopify/react-native-skia, Reanimated, and Gesture Handler. Use when the user wants canvas-driven UI, shaders, paths, rich text, image filters, sprite fields, Skottie, video frames, snapshots, web CanvasKit setup, or performance tuning for custom motion-heavy elements such as loaders, hero art, cards, charts, progress indicators, particle systems, or gesture-driven surfaces. Also use when the user asks for fluid, glow, glass, blob, parallax, 60fps/120fps, or GPU-friendly animated effects in React Native, even if they do not explicitly say "Skia". Do not use for ordinary form/layout work with standard views.
plaid
IncludedProduct Led AI Development — guides founders from idea to launched product. Six capabilities: Idea (discover a product idea), Validate (pressure-test the idea against fatal flaws, problem reality, competition, and 2-week MVP feasibility), Plan (vision intake + document generation), Design (translate image references into a design.md spec), Launch (go-to-market strategy), and Build (roadmap execution). Use when someone says "PLAID", "plaid idea", "help me find an idea", "product idea", "idea from my business", "idea from my expertise", "plaid validate", "validate my idea", "pressure-test", "is this idea good", "find fatal flaws", "validate the problem", "plan a product", "define my vision", "generate a PRD", "product strategy", "plaid design", "design from image", "translate image to design", "create design.md", "extract design tokens", "plaid launch", "go-to-market", "launch plan", "GTM strategy", "launch playbook", "plaid build", "build the app", "start building", or "execute the roadmap".
nextjs-framer-motion-animations
IncludedAdds production-safe Motion for React or Framer Motion animations to Next.js apps, including reveal, hover and tap micro-interactions, whileInView, stagger, AnimatePresence, layout and layoutId transitions, reorder, scroll-linked UI, and lightweight route-content transitions. Use when the user asks to add, refactor, or debug Motion or Framer Motion in App Router or Pages Router codebases, especially around server/client boundaries, reduced motion, LazyMotion, bundle size, hydration, or route transitions. Avoid for GSAP-style timelines, WebGL or 3D scenes, heavy scroll storytelling, or CSS-only effects unless Motion is explicitly requested.