Automator

Automator 技能

Build profitable 自动化工作流 that save hours every week

When to Use

✅ USE this 技能 when:

"Automate my dAIly 报告 generation" "创建 a 工作流 that 监控s prices and alerts me" "设置 up a multi-step data processing 流水线" "I need to schedule recurring tasks with dependencies" "Automate my social media posting across 平台s" "创建 an 应用roval 工作流 for my team" "设置 up automated 备份s with 通知" When NOT to Use

❌ DON'T use this 技能 when:

Single simple command needed (use direct command instead) One-off manual task (no 自动化 needed) Tasks requiring human judgment/creativity Real-time interactive work (工作流 添加s latency) 💰 Value Proposition

What you 获取:

⏰ Save 10+ hours/week on repetitive tasks 🎯 Reliability - 工作流s 运行 on schedule, even when you for获取 🔄 Scalability - same 工作流 works at any volume 📊 Visibility - 追踪 execution 历史 and 失败s 🛡️ Error handling - retries, fallbacks, alerts

ROI Example:

Simple 工作流 (data fetch + emAIl): 1 hour 设置up = 5 hours/month saved Complex 工作流 (multi-source aggregation + 报告s): 4 hours 设置up = 20+ hours/month saved Break-even: 1-2 weeks for most 工作流s Core Concepts 工作流 Structure 工作流: name: "DAIly 报告 生成器" schedule: "0 8 " # Every day at 8 AM steps: - id: fetch_data task: "Fetch sales data from API" 代理: "data-fetcher" timeout: 300

- id: process task: "Process data into 报告 格式化" 代理: "data-处理器" depends_on: [fetch_data] timeout: 600

- id: 通知 task: "发送 报告 via emAIl" 代理: "notifier" depends_on: [process] timeout: 120

代理 角色s

Each step can 运行 on a specialized 代理:

data-fetcher: API calls, data 提取ion data-处理器: Trans格式化ions, analysis, calculations notifier: EmAIl, Slack, Telegram, 通知 应用rover: Human-in-the-loop decisions 归档r: Storage, 备份s, 清理up Error Handling & Retries retry_policy: max_attempts: 3 backoff: "exponential" # 1s, 2s, 4s on_失败: "通知_admin" # or "continue", "abort"

失败_通知: - emAIl: "admin@company.com" - slack: "#alerts"

Quick 启动

• Define Your 工作流

创建 a YAML file my-工作流.yaml:

工作流: name: "Price 监控" description: "检查 product prices hourly and alert if below threshold" schedule: type: "interval" every: "1h"

steps: - name: "检查 Amazon Price" 代理: "price-检查er" prompt: | 检查 price of product https://amazon.com/dp/B08XYZ Return price and avAIlability

- name: "Compare to Threshold" 代理: "decision-maker" prompt: | Threshold: $50 Current price: {{检查 Amazon Price.输出}} Is price below threshold? Return yes/no

- name: "发送 Alert if Cheap" 代理: "notifier" prompt: | If {{Compare to Threshold.输出}} == "yes": 发送 emAIl to user@example.com Subject: Price Alert! Body: Product is now ${{检查 Amazon Price.输出}} depends_on: [检查 Amazon Price, Compare to Threshold]

• Load and 启动

Common 工作流 Patterns Pattern 1: Data 流水线 工作流: name: "DAIly 分析 流水线" schedule: "0 6 " # 6 AM dAIly

steps: - fetch: "提取 data from 3 sources" 代理: "提取器" parallel: true # 运行 multiple sources in parallel

- 转换: "清理 and normalize data" 代理: "转换er" depends_on: [fetch]

- analyze: "生成 insights" 代理: "analyst" depends_on: [转换]

- 报告: "创建 PDF 报告" 代理: "报告er" depends_on: [analyze]

- distribute: "EmAIl and Slack" 代理: "distributor" depends_on: [报告]

Benefit: 30-minute manual process → fully automated

Pattern 2: 应用roval 工作流 工作流: name: "Document 应用roval" trigger: "manual" # 启动 on demand

steps: - draft: "生成 initial document" 代理: "writer"

- review: "Human review" 代理: "应用rover" type: "human_输入" # WAIts for manual 应用roval

- finalize: "应用ly final changes" 代理: "editor" depends_on: [review]

- publish: "部署 to production" 代理: "publisher" depends_on: [finalize]

Benefit: 追踪 应用rovals, no lost emAIls

Pattern 3: Alert & Escalation 工作流: name: "系统 监控" schedule: "/5 *" # Every 5 minutes alert_levels: - 警告: "系统 load > 80%" - critical: "系统 load > 95%"

steps: - 检查: "监控 系统 指标" 代理: "监控"

- classify: "Determine severity" 代理: "classifier"