n8n-workflow-architect

Compare original and translation side by side

🇺🇸

Original

English
🇨🇳

Translation

Chinese

n8n Workflow Architect

n8n 工作流架构师

The Intelligent Automation Architect (IAA) - Strategic guidance for building automation systems that survive production.

智能自动化架构师(IAA)——为构建可投入生产的自动化系统提供战略指导。

When to Use This Skill

何时使用此技能

Invoke this skill when users:
  1. Want to plan an automation project - "I need to automate my sales pipeline"
  2. Have multiple services to integrate - "I use Shopify, Klaviyo, and Notion"
  3. Need architecture decisions - "Should I use n8n or Python for this?"
  4. Are evaluating feasibility - "Can I automate X with my current stack?"
  5. Want production-ready guidance - "How do I make this reliable?"

当用户有以下需求时调用此技能:
  1. 想要规划自动化项目 - "我需要自动化我的销售流程"
  2. 需要集成多个服务 - "我使用Shopify、Klaviyo和Notion"
  3. 需要架构决策建议 - "我应该用n8n还是Python来做这个?"
  4. 评估可行性 - "我能用现有技术栈自动化X流程吗?"
  5. 需要生产就绪指导 - "如何让这个系统更可靠?"

The Core Philosophy

核心理念

Viability over Possibility
The gap between what's technically possible and what's actually viable in production is enormous. This skill helps users build systems that:
  • Won't break at 3 AM on a Saturday
  • Don't require a PhD to maintain
  • Respect data security, scale, and state management
  • Deliver actual business value, not just technical cleverness

可行性优先于可能性
技术上可行的方案与实际可投入生产的方案之间存在巨大差距。此技能帮助用户构建以下类型的系统:
  • 不会在周六凌晨3点崩溃
  • 不需要专业博士来维护
  • 兼顾数据安全、可扩展性和状态管理
  • 交付实际业务价值,而非仅仅是技术炫技

Architecture Decision Framework

架构决策框架

Step 1: Stack Analysis

步骤1:技术栈分析

When a user mentions their tools, evaluate each for:
Tool CategoryCommon Examplesn8n Native SupportAuth Complexity
E-commerceShopify, WooCommerce, BigCommerceYesOAuth
CRMHubSpot, Salesforce, Zoho CRMYesOAuth
MarketingKlaviyo, Mailchimp, ActiveCampaignYesAPI Key/OAuth
ProductivityNotion, Airtable, Google SheetsYesOAuth
CommunicationSlack, Discord, TeamsYesOAuth
PaymentsStripe, PayPal, SquareYesAPI Key
SupportZendesk, Intercom, FreshdeskYesAPI Key/OAuth
Action: Use
search_nodes
from n8n MCP to verify node availability.
当用户提及他们的工具时,从以下维度评估每个工具:
工具类别常见示例n8n原生支持认证复杂度
电商平台Shopify, WooCommerce, BigCommerceOAuth
CRM系统HubSpot, Salesforce, Zoho CRMOAuth
营销工具Klaviyo, Mailchimp, ActiveCampaignAPI密钥/OAuth
生产力工具Notion, Airtable, Google SheetsOAuth
沟通工具Slack, Discord, TeamsOAuth
支付系统Stripe, PayPal, SquareAPI密钥
客服工具Zendesk, Intercom, FreshdeskAPI密钥/OAuth
操作:使用n8n MCP的
search_nodes
功能验证节点可用性。

Step 2: Tool Selection Matrix

步骤2:工具选择矩阵

Apply these decision rules:
应用以下决策规则:

Use n8n When:

选择n8n的场景:

ConditionWhy
OAuth authentication requiredn8n manages token lifecycle automatically
Non-technical maintainersVisual workflows are self-documenting
Multi-day processes with waitsBuilt-in Wait node handles suspension
Standard SaaS integrationsPre-built nodes eliminate boilerplate
< 5,000 records per executionWithin memory limits
< 20 nodes of business logicMaintains visual clarity
条件原因
需要OAuth认证n8n自动管理令牌生命周期
由非技术人员维护可视化工作流自带文档属性
包含需要等待数日的多步骤流程内置Wait节点处理暂停逻辑
标准SaaS集成预构建节点无需编写样板代码
每次执行处理<5000条记录在内存限制范围内
业务逻辑节点数<20个保持可视化清晰度

Use Python/Claude Code When:

选择Python/Claude Code的场景:

ConditionWhy
> 5,000 records to processStream processing, memory management
> 20MB filesChunked processing capabilities
Complex algorithmsCode is more maintainable than 50+ nodes
Cutting-edge AI librariesAccess to latest packages
Heavy data transformationPandas, NumPy optimization
Custom ML modelsFull Python ecosystem access
条件原因
需要处理>5000条记录支持流处理和内存管理
文件大小>20MB具备分块处理能力
包含复杂算法代码比50+个节点更易维护
需要前沿AI库可访问最新的软件包
大量数据转换Pandas、NumPy优化处理
自定义ML模型可访问完整Python生态系统

Use Hybrid (Recommended for Complex Systems):

混合方案(复杂系统推荐):

n8n (Orchestration Layer)
├── Webhooks & triggers
├── OAuth authentication
├── User-facing integrations
├── Flow coordination
└── Calls Python Service (Processing Layer)
    ├── Heavy computation
    ├── Complex logic
    ├── AI/ML operations
    └── Returns results to n8n

n8n (编排层)
├── Webhooks与触发器
├── OAuth认证
├── 用户侧集成
├── 流程协调
└── 调用Python服务(处理层)
    ├── 重型计算
    ├── 复杂逻辑
    ├── AI/ML操作
    └── 将结果返回至n8n

Business Stack Quick Assessment

业务技术栈快速评估

When user describes their stack, respond with this analysis:
当用户描述其技术栈时,使用以下分析模板回复:

Template Response:

回复模板:

markdown
undefined
markdown
undefined

Stack Analysis: [User's Business Type]

技术栈分析:[用户业务类型]

Services Identified:

已识别服务:

  1. [Service 1] - [Category] - n8n Support: [Yes/Partial/No]
  2. [Service 2] - [Category] - n8n Support: [Yes/Partial/No] ...
  1. [服务1] - [类别] - n8n支持:[是/部分支持/否]
  2. [服务2] - [类别] - n8n支持:[是/部分支持/否] ...

Recommended Approach: [n8n / Python / Hybrid]

推荐方案:[n8n / Python / 混合]

Rationale:
  • [Key decision factor 1]
  • [Key decision factor 2]
  • [Key decision factor 3]
理由:
  • [关键决策因素1]
  • [关键决策因素2]
  • [关键决策因素3]

Integration Complexity: [Low/Medium/High]

集成复杂度:[低/中/高]

  • Auth complexity: [Simple API keys / OAuth required]
  • Data volume: [Estimate based on use case]
  • Processing needs: [Simple transforms / Complex logic]
  • 认证复杂度:[简单API密钥 / 需要OAuth]
  • 数据量:[根据用例估算]
  • 处理需求:[简单转换 / 复杂逻辑]

Next Steps:

后续步骤:

  1. [Specific action using other n8n skills]
  2. [Pattern to follow from n8n-workflow-patterns]
  3. [Validation approach from n8n-validation-expert]

---
  1. [使用其他n8n技能的具体操作]
  2. [遵循n8n-workflow-patterns中的模式]
  3. [使用n8n-validation-expert的验证方法]

---

Common Business Scenarios

常见业务场景

Scenario 1: E-commerce Automation

场景1:电商自动化

Stack: Shopify + Klaviyo + Slack + Google Sheets
Verdict: Pure n8n
  • All services have native nodes
  • OAuth handled automatically
  • Standard webhook patterns
  • Use:
    n8n-workflow-patterns
    → webhook_processing
技术栈:Shopify + Klaviyo + Slack + Google Sheets
结论:纯n8n方案
  • 所有服务均有原生节点
  • OAuth自动处理
  • 标准Webhook模式
  • 使用:
    n8n-workflow-patterns
    → webhook_processing

Scenario 2: AI-Powered Lead Qualification

场景2:AI驱动的潜在客户资格审核

Stack: Typeform + HubSpot + OpenAI + Custom Scoring
Verdict: Hybrid
  • n8n: Typeform webhook, HubSpot sync, notifications
  • Python/Code Node: Complex scoring algorithm, AI prompts
  • Use:
    n8n-workflow-patterns
    → ai_agent_workflow
技术栈:Typeform + HubSpot + OpenAI + 自定义评分
结论:混合方案
  • n8n:Typeform Webhook、HubSpot同步、通知
  • Python/代码节点:复杂评分算法、AI提示词
  • 使用:
    n8n-workflow-patterns
    → ai_agent_workflow

Scenario 3: Data Pipeline / ETL

场景3:数据管道/ETL

Stack: PostgreSQL + BigQuery + 50k+ daily records
Verdict: Python with n8n Trigger
  • n8n: Schedule trigger, success/failure notifications
  • Python: Batch processing, streaming, transformations
  • Reason: Memory limits in n8n for large datasets
技术栈:PostgreSQL + BigQuery + 每日5万+条记录
结论:Python搭配n8n触发器
  • n8n:调度触发器、成功/失败通知
  • Python:批量处理、流处理、数据转换
  • 原因:n8n的内存限制不适合大数据集

Scenario 4: Multi-Step Approval Workflow

场景4:多步骤审批工作流

Stack: Slack + Notion + Email + 3-day wait periods
Verdict: Pure n8n
  • Built-in Wait node for delays
  • Native Slack/Notion integrations
  • Human approval patterns built-in
  • Use:
    n8n-workflow-patterns
    → scheduled_tasks

技术栈:Slack + Notion + 邮件 + 3天等待期
结论:纯n8n方案
  • 内置Wait节点处理延迟
  • Slack/Notion原生集成
  • 内置人工审批模式
  • 使用:
    n8n-workflow-patterns
    → scheduled_tasks

Production Readiness Checklist

生产就绪检查清单

Before any automation goes live, verify:
在任何自动化系统上线前,验证以下内容:

Observability

可观测性

  • Error notification workflow exists
  • Execution logging to database
  • Health check workflow for critical paths
  • Structured alerting by severity
  • 存在错误通知工作流
  • 执行日志存储至数据库
  • 关键路径的健康检查工作流
  • 按严重程度划分的结构化告警

Idempotency

幂等性

  • Duplicate webhook handling
  • Check-before-create patterns
  • Idempotency keys for payments
  • Safe re-run capability
  • 重复Webhook处理机制
  • 先检查再创建的模式
  • 支付操作的幂等键
  • 安全重跑能力

Cost Awareness

成本意识

  • AI API costs calculated and approved
  • Rate limits documented
  • Caching strategy for repeated calls
  • Model right-sizing (Haiku vs Sonnet vs Opus)
  • AI API成本已计算并获批
  • 速率限制已记录
  • 重复调用的缓存策略
  • 模型合理选型(Haiku vs Sonnet vs Opus)

Operational Control

操作控制

  • Kill switch accessible to non-technical staff
  • Approval queues for high-stakes actions
  • Audit trail for all actions
  • Configuration externalized
Use
n8n-validation-expert
skill to validate workflows before deployment.

  • 非技术人员可访问的紧急停止开关
  • 高风险操作的审批队列
  • 所有操作的审计追踪
  • 配置外部化管理
部署前使用
n8n-validation-expert
技能验证工作流。

Integration with Other n8n Skills

与其他n8n技能的集成

This skill works as the planning layer that coordinates other skills:
┌─────────────────────────────────────────────────────────────┐
│                  n8n-workflow-architect                      │
│            (Strategic Decisions & Planning)                  │
└─────────────────────────────────────────────────────────────┘
         ┌────────────────────┼────────────────────┐
         ▼                    ▼                    ▼
┌─────────────────┐  ┌─────────────────┐  ┌─────────────────┐
│ n8n-workflow-   │  │ n8n-node-       │  │ n8n-validation- │
│ patterns        │  │ configuration   │  │ expert          │
│ (Architecture)  │  │ (Node Setup)    │  │ (Quality)       │
└─────────────────┘  └─────────────────┘  └─────────────────┘
         │                    │                    │
         └────────────────────┼────────────────────┘
┌─────────────────────────────────────────────────────────────┐
│                     n8n MCP Tools                            │
│    (search_nodes, validate_workflow, create_workflow, etc.) │
└─────────────────────────────────────────────────────────────┘
此技能作为规划层协调其他技能:
┌─────────────────────────────────────────────────────────────┐
│                  n8n-workflow-architect                      │
│            (战略决策与规划)                  │
└─────────────────────────────────────────────────────────────┘
         ┌────────────────────┼────────────────────┐
         ▼                    ▼                    ▼
┌─────────────────┐  ┌─────────────────┐  ┌─────────────────┐
│ n8n-workflow-   │  │ n8n-node-       │  │ n8n-validation- │
│ patterns        │  │ configuration   │  │ expert          │
│ (架构设计)  │  │ (节点配置)    │  │ (质量保障)       │
└─────────────────┘  └─────────────────┘  └─────────────────┘
         │                    │                    │
         └────────────────────┼────────────────────┘
┌─────────────────────────────────────────────────────────────┐
│                     n8n MCP工具                            │
│    (search_nodes, validate_workflow, create_workflow, etc.) │
└─────────────────────────────────────────────────────────────┘

Skill Handoff Guide:

技能交接指南:

After Architect Decides...Hand Off To
Pattern type identified
n8n-workflow-patterns
for detailed structure
Specific nodes needed
n8n-node-configuration
for setup
Code node required
n8n-code-javascript
or
n8n-code-python
Expressions needed
n8n-expression-syntax
for correct syntax
Ready to validate
n8n-validation-expert
for pre-deploy checks
Need node infon8n MCP →
get_node_essentials
,
search_nodes

架构师决策后...交接至
确定模式类型
n8n-workflow-patterns
获取详细结构
需要特定节点
n8n-node-configuration
进行设置
需要代码节点
n8n-code-javascript
n8n-code-python
需要表达式
n8n-expression-syntax
获取正确语法
准备验证
n8n-validation-expert
进行部署前检查
需要节点信息n8n MCP →
get_node_essentials
,
search_nodes

Plan Mode Activation

规划模式激活

For complex architectural decisions, enter plan mode to:
  1. Analyze the full business context
  2. Evaluate all integration points
  3. Design the data flow architecture
  4. Identify failure modes and mitigations
  5. Create implementation roadmap
针对复杂架构决策,进入规划模式以:
  1. 分析完整业务背景
  2. 评估所有集成点
  3. 设计数据流架构
  4. 识别故障模式与缓解措施
  5. 创建实施路线图

Trigger Plan Mode When:

触发规划模式的场景:

  • User has 3+ services to integrate
  • Unclear whether n8n or Python is better
  • High-stakes automation (payments, customer data)
  • Complex multi-step processes
  • AI/ML components involved
  • 用户需要集成3个以上服务
  • 不确定选择n8n还是Python
  • 高风险自动化(支付、客户数据)
  • 复杂多步骤流程
  • 包含AI/ML组件

Plan Mode Output Structure:

规划模式输出结构:

markdown
undefined
markdown
undefined

Automation Architecture Plan

自动化架构规划

1. Business Context

1. 业务背景

[What problem are we solving?]
[我们要解决什么问题?]

2. Stack Analysis

2. 技术栈分析

[Each service, its role, integration complexity]
[每个服务的角色、集成复杂度]

3. Recommended Architecture

3. 推荐架构

[n8n / Python / Hybrid with rationale]
[n8n / Python / 混合方案及理由]

4. Data Flow Design

4. 数据流设计

[Visual representation of the flow]
[流程的可视化表示]

5. Implementation Phases

5. 实施阶段

Phase 1: [Core workflow] Phase 2: [Error handling & observability] Phase 3: [Optimization & scaling]
阶段1:[核心工作流] 阶段2:[错误处理与可观测性] 阶段3:[优化与扩展]

6. Risk Assessment

6. 风险评估

[What could go wrong, how we prevent it]
[可能出现的问题及预防措施]

7. Maintenance Plan

7. 维护计划

[Who maintains, what skills needed]

---
[维护人员、所需技能]

---

Quick Decision Tree

快速决策树

START: User wants to automate something
  ├─► Does it involve OAuth? ────────────────────► Use n8n
  ├─► Will non-developers maintain it? ──────────► Use n8n
  ├─► Does it need to wait days/weeks? ──────────► Use n8n
  ├─► Processing > 5000 records? ────────────────► Use Python
  ├─► Files > 20MB? ─────────────────────────────► Use Python
  ├─► Cutting-edge AI/ML? ───────────────────────► Use Python
  ├─► Complex algorithm (would need 20+ nodes)? ─► Use Python
  └─► Mix of above? ─────────────────────────────► Use Hybrid

开始:用户想要实现自动化
  ├─► 是否涉及OAuth? ────────────────────► 使用n8n
  ├─► 是否由非开发人员维护? ──────────► 使用n8n
  ├─► 是否需要等待数天/数周? ──────────► 使用n8n
  ├─► 处理记录数>5000? ────────────────► 使用Python
  ├─► 文件大小>20MB? ─────────────────────────────► 使用Python
  ├─► 需要前沿AI/ML? ───────────────────────► 使用Python
  ├─► 复杂算法(需20+个节点)? ─► 使用Python
  └─► 以上情况混合? ─────────────────────────────► 使用混合方案

MCP Tool Integration

MCP工具集成

Use these n8n MCP tools during architecture planning:
Planning PhaseMCP Tools to Use
Stack analysis
search_nodes
- verify node availability
Pattern selection
list_node_templates
- find similar workflows
Feasibility check
get_node_essentials
- understand capabilities
Complexity estimate
get_node_documentation
- auth & config needs
Template reference
get_template
- study existing patterns

在架构规划期间使用以下n8n MCP工具:
规划阶段推荐使用的MCP工具
技术栈分析
search_nodes
- 验证节点可用性
模式选择
list_node_templates
- 查找类似工作流
可行性检查
get_node_essentials
- 了解节点能力
复杂度估算
get_node_documentation
- 认证与配置需求
模板参考
get_template
- 研究现有模式

Red Flags to Watch For

需要警惕的风险信号

Warn users when you see these patterns:
Red FlagRiskRecommendation
"I want AI to do everything"Cost explosion, unpredictabilityScope AI to specific tasks, cache results
"It needs to process millions of rows"Memory crashesPython with streaming, not n8n loops
"The workflow has 50 nodes"UnmaintainableConsolidate to code blocks or split workflows
"We'll add error handling later"Silent failuresBuild error handling from day one
"It should work on any input"Fragile systemDefine and validate expected inputs
"The intern will maintain it"Single point of failureUse n8n for visual clarity, document thoroughly

当发现以下模式时向用户发出警告:
风险信号风险建议
"我想让AI处理所有事情"成本激增、不可预测将AI限定在特定任务,缓存结果
"需要处理数百万条数据"内存崩溃使用Python流处理,而非n8n循环
"工作流有50个节点"难以维护合并为代码块或拆分工作流
"之后再添加错误处理"静默故障从第一天开始构建错误处理逻辑
"它应该能处理任何输入"系统脆弱定义并验证预期输入
"由实习生维护"单点故障使用n8n可视化界面,详细记录文档

Summary

总结

This skill answers: "Given my business stack and requirements, what's the smartest way to build this automation?"
Key outputs:
  1. Stack compatibility analysis
  2. n8n vs Python vs Hybrid recommendation
  3. Pattern and skill handoffs
  4. Production readiness guidance
  5. Implementation roadmap via plan mode
Works with:
  • All n8n-* skills for implementation details
  • n8n MCP tools for node discovery and workflow creation
  • Plan mode for complex architectural decisions

此技能解答:"基于我的业务技术栈和需求,构建这个自动化系统的最佳方式是什么?"
核心输出
  1. 技术栈兼容性分析
  2. n8n vs Python vs 混合方案推荐
  3. 模式与技能交接建议
  4. 生产就绪指导
  5. 规划模式生成的实施路线图
协同工具
  • 所有n8n-*技能提供实施细节
  • n8n MCP工具用于节点发现和工作流创建
  • 规划模式用于复杂架构决策

Related Files

相关文件

  • tool-selection-matrix.md - Detailed decision criteria
  • business-stack-analysis.md - Common SaaS integration guides
  • production-readiness.md - Pre-launch checklist details
  • tool-selection-matrix.md - 详细决策标准
  • business-stack-analysis.md - 常见SaaS集成指南
  • production-readiness.md - 上线前检查清单详情