Back to Details

agent-workflow-designer

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

Agent Workflow Designer

Agent工作流设计器

Tier: POWERFUL
Category: Engineering
Domain: Multi-Agent Systems / AI Orchestration
层级： 强大级
分类： 工程类
领域： 多Agent系统 / AI编排

Overview

概述

Design production-grade multi-agent workflows with clear pattern choice, handoff contracts, failure handling, and cost/context controls.
设计生产级别的多Agent工作流，包含清晰的模式选择、交接协议、故障处理以及成本/上下文控制。

Core Capabilities

核心功能

  • Workflow pattern selection for multi-step agent systems
  • Skeleton config generation for fast workflow bootstrapping
  • Context and cost discipline across long-running flows
  • Error recovery and retry strategy scaffolding
  • Documentation pointers for operational pattern tradeoffs
  • 为多步骤Agent系统选择工作流模式
  • 生成骨架配置以快速启动工作流
  • 在长期运行的流程中管控上下文与成本
  • 搭建错误恢复与重试策略框架
  • 提供操作模式权衡相关的文档指引

When to Use

使用场景

  • A single prompt is insufficient for task complexity
  • You need specialist agents with explicit boundaries
  • You want deterministic workflow structure before implementation
  • You need validation loops for quality or safety gates
  • 单个提示词不足以应对任务复杂度
  • 需要具备明确边界的专业Agent
  • 希望在实现前确定工作流的确定性结构
  • 需要用于质量或安全校验的验证循环

Quick Start

快速开始

bash
undefined
bash
undefined

Generate a sequential workflow skeleton

Generate a sequential workflow skeleton

python3 scripts/workflow_scaffolder.py sequential --name content-pipeline
python3 scripts/workflow_scaffolder.py sequential --name content-pipeline

Generate an orchestrator workflow and save it

Generate an orchestrator workflow and save it

python3 scripts/workflow_scaffolder.py orchestrator --name incident-triage --output workflows/incident-triage.json

---
python3 scripts/workflow_scaffolder.py orchestrator --name incident-triage --output workflows/incident-triage.json

---

Pattern Map

模式图谱

  • sequential
    : strict step-by-step dependency chain
  • parallel
    : fan-out/fan-in for independent subtasks
  • router
    : dispatch by intent/type with fallback
  • orchestrator
    : planner coordinates specialists with dependencies
  • evaluator
    : generator + quality gate loop
Detailed templates: 
references/workflow-patterns.md
  • sequential
    : 严格的分步依赖链
  • parallel
    : 用于独立子任务的扇出/扇入模式
  • router
    : 按意图/类型分配任务并提供回退机制
  • orchestrator
    : 规划器协调带有依赖关系的专业Agent
  • evaluator
    : 生成器+质量校验循环
详细模板：
references/workflow-patterns.md

Recommended Workflow

推荐工作流

  1. Select pattern based on dependency shape and risk profile.
  2. Scaffold config via 
    scripts/workflow_scaffolder.py
    .
  3. Define handoff contract fields for every edge.
  4. Add retry/timeouts and output validation gates.
  5. Dry-run with small context budgets before scaling.
  1. 根据依赖形态和风险概况选择模式。
  2. 通过
    scripts/workflow_scaffolder.py
    搭建配置骨架。
  3. 为每个节点定义交接协议字段。
  4. 添加重试/超时机制和输出校验门限。
  5. 在扩容前使用小上下文预算进行试运行。

Common Pitfalls

常见误区

  • Over-orchestrating tasks solvable by one well-structured prompt
  • Missing timeout/retry policies for external-model calls
  • Passing full upstream context instead of targeted artifacts
  • Ignoring per-step cost accumulation
  • 过度编排可通过单个结构良好的提示词解决的任务
  • 外部模型调用缺少超时/重试策略
  • 传递完整上游上下文而非针对性的工件
  • 忽略每一步的成本累积

Best Practices

最佳实践

  1. Start with the smallest pattern that can satisfy requirements.
  2. Keep handoff payloads explicit and bounded.
  3. Validate intermediate outputs before fan-in synthesis.
  4. Enforce budget and timeout limits in every step.
  1. 从能满足需求的最小模式开始。
  2. 保持交接负载明确且可控。
  3. 在扇入合成前验证中间输出。
  4. 在每一步强制实施预算和超时限制。