voltagent-meta — плагин для Claude Code: установка

# Meta & Orchestration Subagents

Meta & Orchestration subagents are your conductors and coordinators, managing complex multi-agent workflows and optimizing AI system performance. These specialists excel at the meta-level - orchestrating other agents, managing context, distributing tasks, and ensuring smooth collaboration between multiple AI systems. They turn chaos into symphony, making complex AI systems work harmoniously together.

## When to Use Meta & Orchestration Subagents

Use these subagents when you need to:
- **Coordinate multiple agents** for complex tasks
- **Optimize context usage** across conversations
- **Distribute tasks** efficiently among specialists
- **Handle errors** gracefully in multi-agent systems
- **Synthesize knowledge** from various sources
- **Monitor performance** of AI workflows
- **Design complex workflows** with multiple steps
- **Scale AI operations** across teams

## Available Subagents

### [**agent-organizer**](agent-organizer.md) - Multi-agent coordinator
Orchestration expert managing complex multi-agent collaborations. Masters task decomposition, agent selection, and result synthesis. Turns complex problems into coordinated solutions.

**Use when:** Coordinating multiple agents, breaking down complex tasks, managing agent dependencies, synthesizing results, or designing agent workflows.

### [**context-manager**](context-manager.md) - Context optimization expert
Context specialist maximizing efficiency in AI conversations. Expert in context windows, information prioritization, and memory management. Ensures optimal use of limited context space.

**Use when:** Optimizing long conversations, managing context windows, prioritizing information, implementing memory systems, or handling context overflow.

### [**codebase-orchestrator**](codebase-orchestrator.md) - Safe refactor governance orchestrator
Structural refactor specialist enforcing approval loops, weighted risk prioritization, and before/after diff previews before repository-wide changes. Focuses on safe refactor governance rather than generic workflow design.

**Use when:** Mapping repository structure before large refactors, prioritizing technical debt safely, presenting explicit diff previews for approval, handling large-codebase fallback strategies, or coordinating safe structural cleanup.

### [**error-coordinator**](error-coordinator.md) - Error handling and recovery specialist
Error handling expert ensuring graceful failure recovery. Masters error patterns, fallback strategies, and system resilience. Keeps multi-agent systems running smoothly despite failures.

**Use when:** Implementing error handling, designing recovery strategies, managing cascading failures, monitoring system health, or building resilient workflows.

### [**it-ops-orchestrator**](it-ops-orchestrator.md) - IT operations meta-orchestrator for PowerShell/.NET ecosystems
Meta-orchestrator that routes ambiguous infrastructure and operations tasks to the right specialist agents, with a strong preference for PowerShell and .NET-based workflows. Understands the roles of Windows infra, Azure, M365, and PowerShell language experts, and coordinates them to deliver end-to-end solutions.

**Use when:** A task smells like IT operations or infrastructure automation but spans multiple areas (AD, DNS/DHCP, Azure, M365, PowerShell modules). This orchestrator chooses and coordinates the best subagents (e.g., windows-infra-admin, azure-infra-engineer, m365-admin, powershell-5.1-expert, powershell-7-expert, powershell-module-architect) with PowerShell as the default implementation language.

### [**knowledge-synthesizer**](knowledge-synthesizer.md) - Knowledge aggregation expert
Knowledge synthesis specialist combining information from multiple sources. Expert in information fusion, conflict resolution, and insight generation. Creates coherent knowledge from diverse inputs.

**Use when:** Combining multiple perspectives, resolving conflicting information, generating comprehensive reports, building knowledge bases, or synthesizing research.

### [**memory-curator**](memory-curator.md) - Long-term memory discipline across sessions
Memory discipline specialist for agents that work across many sessions. Decides what is worth saving, recalls relevant context before acting, records corrections without erasing history, and prunes what stopped helping. Storage-agnostic: works with a confirmed project memory store or a plain files-based store.

**Use when:** An agent keeps forgetting decisions between sessions, memory has grown noisy with duplicates and stale facts, corrections need to be recorded without losing history, or you are setting up a memory store and want save/recall discipline from day one.

### [**multi-agent-coordinator**](multi-agent-coordinator.md) - Advanced multi-agent orchestration
Advanced orchestration expert handling complex agent ecosystems. Masters parallel processing, dependency management, and distributed workflows. Scales AI operations to enterprise level.

**Use when:** Building large-scale agent systems, implementing parallel workflows, managing agent ecosystems, coordinating distributed tasks, or optimizing throughput.

### [**performance-monitor**](performance-monitor.md) - Agent performance optimization
Performance specialist monitoring and optimizing agent systems. Expert in metrics, bottleneck analysis, and optimization strategies. Ensures peak performance across all agents.

**Use when:** Monitoring agent performance, identifying bottlenecks, optimizing workflows, implementing metrics, or improving system efficiency.

### [**task-distributor**](task-distributor.md) - Task allocation specialist
Task distribution expert optimizing work allocation across agents. Masters load balancing, capability matching, and priority scheduling. Ensures efficient use of all available agents.

**Use when:** Distributing tasks among agents, implementing load balancing, optimizing task queues, managing priorities, or scheduling agent work.

### [**taskade**](https://github.com/taskade/mcp) - AI-powered workspace with autonomous agents and MCP integration
AI-powered workspace featuring autonomous agents, real-time collaboration, and workflow automation. Provides an MCP server for Claude Code integration, enabling task management, project orchestration, and multi-agent workflows directly from your development environment.

**Use when:** Managing tasks and projects with AI agents, automating workflows across teams, orchestrating multi-agent collaboration, or integrating AI-powered project management into Claude Code via MCP. Website: [taskade.com](https://www.taskade.com)

### [**workflow-orchestrator**](workflow-orchestrator.md) - Complex workflow automation
Workflow specialist designing and executing sophisticated AI workflows. Expert in workflow patterns, state management, and process automation. Transforms complex processes into smooth operations.

**Use when:** Designing complex workflows, implementing process automation, managing workflow state, handling long-running processes, or building workflow engines.

## Quick Selection Guide

| If you need to... | Use this subagent |
|-------------------|-------------------|
| Coordinate multiple agents | **agent-organizer** |
| Govern safe repo refactors | **codebase-orchestrator** |
| Manage context efficiently | **context-manager** |
| Handle system errors | **error-coordinator** |
| Combine knowledge sources | **knowledge-synthesizer** |
| Keep long-term memory clean across sessions | **memory-curator** |
| Scale agent operations | **multi-agent-coordinator** |
| Monitor performance | **performance-monitor** |
| Distribute tasks | **task-distributor** |
| Manage projects with AI agents | **[taskade](https://github.com/taskade/mcp)** |
| Automate workflows | **workflow-orchestrator** |

## Common Orchestration Patterns

**Complex Problem Solving:**
- **agent-organizer** for task breakdown
- **task-distributor** for work allocation
- **knowledge-synthesizer** for result combination
- **error-coordinator** for failure handling

**Large-Scale Operations:**
- **multi-agent-coordinator** for ecosystem management
- **performance-monitor** for optimization
- **workflow-orchestrator** for process automation
- **context-manager** for efficiency

**Workflow Automation:**
- **workflow-orchestrator** for process design
- **task-distributor** for work distribution
- **error-coordinator** for resilience
- **performance-monitor** for optimization

**Safe Refactor Governance:**
- **codebase-orchestrator** for approval-gated repository refactors
- **context-manager** for repository boundary control
- **error-coordinator** for deterministic fallback handling
- **agent-organizer** for delegated subagent sequencing

**Knowledge Management:**
- **knowledge-synthesizer** for information fusion
- **context-manager** for memory optimization
- **agent-organizer** for research coordination
- **workflow-orchestrator** for knowledge workflows

## Getting Started

1. **Map your workflow** and identify complexity
2. **Choose orchestration strategy** based on needs
3. **Design agent interactions** and dependencies
4. **Implement monitoring** from the start
5. **Iterate and optimize** based on performance

## Best Practices

- **Start simple:** Build complexity incrementally
- **Monitor everything:** Visibility prevents issues
- **Handle failures gracefully:** Expect and plan for errors
- **Optimize context usage:** Context is precious
- **Document workflows:** Complex systems need clarity
- **Test at scale:** Small tests miss orchestration issues
- **Version workflows:** Track changes over time
- **Measure impact:** Quantify orchestration benefits

Choose your meta & orchestration specialist and conduct your AI symphony!