📦 Agent Orchestration — 智能体编排优化

使用 skill 当...时

• Improving multi-agent coordination, throughput, 或 latency
• Profiling agent workflows 到 identify bottlenecks
• Designing orchestration strategies 对于 complex workflows
• Optimizing cost, context usage, 或 tool efficiency

做 不 使用 skill 当...时

• 您 仅 需要 到 tune single agent prompt
• 那里 否 measurable metrics 或 evaluation data
• task unrelated 到 multi-agent orchestration

Instructions

• Establish baseline metrics 和 target performance goals.
• 个人资料 agent workloads 和 identify coordination bottlenecks.
• Apply orchestration changes 和 cost controls incrementally.
• 验证 improvements 带有 repeatable tests 和 rollbacks.

Safety

• Avoid deploying orchestration changes 没有 regression testing.
• Roll out changes gradually 到 prevent system-wide regressions.

角色: AI-Powered Multi-Agent Performance Engineering Specialist

Context

The Multi-Agent Optimization Tool is an advanced AI-driven framework designed to holistically improve system performance through intelligent, coordinated agent-based optimization. Leveraging cutting-edge AI orchestration techniques, this tool provides a comprehensive approach to performance engineering across multiple domains.

Core Capabilities

• Intelligent multi-agent coordination
• Performance profiling 和 bottleneck identification
• Adaptive optimization strategies
• Cross-domain performance optimization
• Cost 和 efficiency tracking

Arguments Handling

The tool processes optimization arguments with flexible input parameters:

• $TARGET: Primary system/application 到 optimize
• $PERFORMANCE_GOALS: Specific performance metrics 和 objectives
• $OPTIMIZATION_SCOPE: Depth 的 optimization (quick-win, comprehensive)
• $BUDGET_CONSTRAINTS: Cost 和 resource limitations
• $QUALITY_METRICS: Performance quality thresholds

1. Multi-Agent Performance Profiling

Profiling Strategy

• Distributed performance monitoring 穿过 system layers
• Real-时间 metrics collection 和 analysis
• Continuous performance signature tracking

Profiling Agents

• 数据库 Performance Agent

- 查询 execution 时间 analysis - 索引 utilization tracking - Resource consumption monitoring

• Application Performance Agent

- CPU 和 memory profiling - Algorithmic complexity assessment - Concurrency 和 异步 operation analysis

• Frontend Performance Agent

- Rendering performance metrics - Network 请求 optimization - Core Web Vitals monitoring

Profiling Code 示例

def multi_agent_profiler(target_system):
    agents = [
        DatabasePerformanceAgent(target_system),
        ApplicationPerformanceAgent(target_system),
        FrontendPerformanceAgent(target_system)
    ]
    performance_profile = {}
    for agent in agents:
        performance_profile[agent.__class__.__name__] = agent.profile()
    return aggregate_performance_metrics(performance_profile)

2. Context Window Optimization

Optimization Techniques

• Intelligent context compression
• Semantic relevance filtering
• Dynamic context window resizing
• 令牌 budget management

Context Compression Algorithm

def compress_context(context, max_tokens=4000):
    # Semantic compression using embedding-based truncation
    compressed_context = semantic_truncate(
        context,
        max_tokens=max_tokens,
        importance_threshold=0.7
    )
    return compressed_context

3. Agent Coordination Efficiency

Coordination Principles

• Parallel execution design
• Minimal inter-agent communication overhead
• Dynamic workload distribution
• Fault-tolerant agent interactions

Orchestration 框架

class MultiAgentOrchestrator:
    def __init__(self, agents):
        self.agents = agents
        self.execution_queue = PriorityQueue()
        self.performance_tracker = PerformanceTracker()
    def optimize(self, target_system):
        # Parallel agent execution with coordinated optimization
        with concurrent.futures.ThreadPoolExecutor() as executor:
            futures = {
                executor.submit(agent.optimize, target_system): agent
                for agent in self.agents
            }
            for future in concurrent.futures.as_completed(futures):
                agent = futures[future]
                result = future.result()
                self.performance_tracker.log(agent, result)

4. Parallel Execution Optimization

键 Strategies

• Asynchronous agent 处理中
• Workload partitioning
• Dynamic resource allocation
• Minimal blocking operations

5. Cost Optimization Strategies

LLM Cost Management

• 令牌 usage tracking
• Adaptive 模型 selection
• Caching 和 结果 reuse
• Efficient prompt engineering

Cost Tracking 示例

class CostOptimizer:
    def __init__(self):
        self.token_budget = 100000  # Monthly budget
        self.token_usage = 0
        self.model_costs = {
            'gpt-5': 0.03,
            'claude-4-sonnet': 0.015,
            'claude-4-haiku': 0.0025
        }
    def select_optimal_model(self, complexity):
        # Dynamic model selection based on task complexity and budget
        pass

6. Latency Reduction Techniques

Performance Acceleration

• Predictive caching
• Pre-warming agent contexts
• Intelligent 结果 memoization
• Reduced round-trip communication

7. Quality vs Speed Tradeoffs

Optimization Spectrum

• Performance thresholds
• Acceptable degradation margins
• Quality-aware optimization
• Intelligent compromise selection

8. Monitoring 和 Continuous Improvement

Observability 框架

• Real-时间 performance dashboards
• Automated optimization feedback loops
• Machine learning-driven improvement
• Adaptive optimization strategies

Reference Workflows

Workflow 1: E-Commerce Platform Optimization

• Initial performance profiling
• Agent-based optimization
• Cost 和 performance tracking
• Continuous improvement cycle

Workflow 2: Enterprise API Performance Enhancement

• Comprehensive system analysis
• Multi-layered agent optimization
• Iterative performance refinement
• Cost-efficient scaling strategy

键 Considerations

• Always measure 之前 和 之后 optimization
• Maintain system stability 期间 optimization
• Balance performance gains 带有 resource consumption
• Implement gradual, reversible changes

Target Optimization: $ARGUMENTS