Konsep Inti Alur Kerja Microsoft Agent Frameflows - Edges

Dokumen ini memberikan tampilan mendalam pada komponen Edges dari sistem Microsoft Agent Framework Workflow.

Gambaran Umum

Tepi menentukan bagaimana pesan mengalir di antara pelaksana dengan kondisi opsional. Mereka mewakili koneksi dalam grafik alur kerja dan menentukan jalur aliran data.

Jenis Tepi

Kerangka kerja mendukung beberapa pola tepi:

1. Tepi Langsung: Koneksi satu-ke-satu sederhana antara para eksekutor
2. Tepi Bersyarat: Tepi dengan kondisi yang menentukan kapan pesan akan mengalir
3. Sisi Fan-out: Satu pemroses mengirim pesan ke beberapa target
4. Fan-in Edges: Beberapa eksekutor mengirim pesan ke satu target

Tepi Langsung

Bentuk koneksi paling sederhana antara dua pelaksana:

using Microsoft.Agents.AI.Workflows;

WorkflowBuilder builder = new(sourceExecutor);
builder.AddEdge(sourceExecutor, targetExecutor);

from agent_framework import WorkflowBuilder

builder = WorkflowBuilder()
builder.add_edge(source_executor, target_executor)
builder.set_start_executor(source_executor)
workflow = builder.build()

Simpul Bersyarat

Tepi yang hanya diaktifkan saat kondisi tertentu terpenuhi:

// Route based on message content
builder.AddEdge(
    source: spamDetector,
    target: emailProcessor,
    condition: result => result is SpamResult spam && !spam.IsSpam
);

builder.AddEdge(
    source: spamDetector,
    target: spamHandler,
    condition: result => result is SpamResult spam && spam.IsSpam
);

from agent_framework import WorkflowBuilder

builder = WorkflowBuilder()
builder.add_edge(spam_detector, email_processor, condition=lambda result: isinstance(result, SpamResult) and not result.is_spam)
builder.add_edge(spam_detector, spam_handler, condition=lambda result: isinstance(result, SpamResult) and result.is_spam)
builder.set_start_executor(spam_detector)
workflow = builder.build()

Tepi switch-case

Merutekan pesan ke pelaksana yang berbeda berdasarkan kondisi:

builder.AddSwitch(routerExecutor, switchBuilder =>
    switchBuilder
        .AddCase(
            message => message.Priority < Priority.Normal,
            executorA
        )
        .AddCase(
            message => message.Priority < Priority.High,
            executorB
        )
        .SetDefault(executorC)
);

from agent_framework import (
    Case,
    Default,
    WorkflowBuilder,
)

builder = WorkflowBuilder()
builder.set_start_executor(router_executor)
builder.add_switch_case_edge_group(
    router_executor,
    [
        Case(
            condition=lambda message: message.priority < Priority.NORMAL,
            target=executor_a,
        ),
        Case(
            condition=lambda message: message.priority < Priority.HIGH,
            target=executor_b,
        ),
        Default(target=executor_c)
    ],
)
workflow = builder.build()

Tepi Fan-out

Distribusikan pesan dari satu pelaksana ke beberapa target:

// Send to all targets
builder.AddFanOutEdge(splitterExecutor, targets: [worker1, worker2, worker3]);

// Send to specific targets based on target selector function
builder.AddFanOutEdge(
    source: routerExecutor,
    targetSelector: (message, targetCount) => message.Priority switch
    {
        Priority.High => [0], // Route to first worker only
        Priority.Normal => [1, 2], // Route to workers 2 and 3
        _ => Enumerable.Range(0, targetCount) // Route to all workers
    },
    targets: [highPriorityWorker, normalWorker1, normalWorker2]
);

from agent_framework import WorkflowBuilder

builder = WorkflowBuilder()
builder.set_start_executor(splitter_executor)
builder.add_fan_out_edges(splitter_executor, [worker1, worker2, worker3])
workflow = builder.build()

# Send to specific targets based on partitioner function
builder = WorkflowBuilder()
builder.set_start_executor(splitter_executor)
builder.add_fan_out_edges(
    splitter_executor,
    [worker1, worker2, worker3],
    selection_func=lambda message, target_ids: (
        [0] if message.priority == Priority.HIGH else
        [1, 2] if message.priority == Priority.NORMAL else
        list(range(target_count))
    )
)
workflow = builder.build()

Tepi Fan-in

Kumpulkan pesan dari beberapa sumber ke dalam satu target:

// Aggregate results from multiple workers
builder.AddFanInEdge(aggregatorExecutor, sources: [worker1, worker2, worker3]);

builder.add_fan_in_edge([worker1, worker2, worker3], aggregator_executor)

Langkah Berikutnya

• Pelajari tentang Alur Kerja untuk memahami cara membuat dan menjalankan alur kerja.