Search news using Perplexity AI and post to X (Twitter)

工作流概述

这是一个包含5个节点的中等工作流,主要用于自动化处理各种任务。

工作流源代码

下载
{
  "id": "v9K61fCQhrG6gt6Z",
  "meta": {
    "instanceId": "9219ebc7795bea866f70aa3d977d54417fdf06c41944be95e20cfb60f992db19",
    "templateCredsSetupCompleted": true
  },
  "name": "Search news using Perplexity AI and post to X (Twitter)",
  "tags": [],
  "nodes": [
    {
      "id": "9b2fdc0f-8c71-4ea9-a9d0-df470f2778df",
      "name": "Schedule Trigger",
      "type": "n8n-nodes-base.scheduleTrigger",
      "position": [
        -560,
        0
      ],
      "parameters": {
        "rule": {
          "interval": [
            {
              "field": "hours",
              "hoursInterval": 21,
              "triggerAtMinute": "={{Math.floor(Math.random() * 60)}}
"
            }
          ]
        }
      },
      "typeVersion": 1.2
    },
    {
      "id": "d549c019-1f3e-4758-a5ee-e4ac7e03cc2d",
      "name": "searchQuery",
      "type": "n8n-nodes-base.set",
      "position": [
        -340,
        0
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "4cfbc312-5dcd-444d-ae08-0bab417c084c",
              "name": "searchInput",
              "type": "string",
              "value": "What's the latest news in artificial intelligence?"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "87c28d43-6cd6-4292-a563-a9f22467e162",
      "name": "Perplexity",
      "type": "n8n-nodes-base.httpRequest",
      "position": [
        100,
        0
      ],
      "parameters": {
        "url": "https://api.perplexity.ai/chat/completions",
        "method": "POST",
        "options": {},
        "jsonBody": "={
  \"model\": \"llama-3.1-sonar-small-128k-online\",
  \"messages\": [
    {
      \"role\": \"system\",
      \"content\": \"You are a social media assistant summarizing tech news for Twitter/X. Only return one article. Your output must follow this exact format: a short, engaging headline (max 140 characters), followed by a single space, then the direct article link. Do not use markdown, hashtags, emojis, or line breaks. Keep the total output under 200 characters. Be precise, objective, and newsworthy.Example: Mastercard launches Agent Pay, allowing AI agents to make purchases for users. https://www.perplexity.ai/page/mastercard-unveils-agent-pay-e-qWXnaUEzQZWCqsxF4l43zA\"
    },
    {
      \"role\": \"user\",
      \"content\": \"{{ $('searchQuery').item.json.searchInput }}\"
    }
  ],
  \"temperature\": 0.3,
  \"top_p\": 0.9,
  \"return_citations\": true,
  \"search_domain_filter\": [
    \"perplexity.ai\"
  ],
  \"search_recency_filter\": \"day\",
  \"return_images\": true,
  \"return_related_questions\": false,
  \"max_tokens\": 80,
  \"stream\": false,
  \"presence_penalty\": 0,
  \"frequency_penalty\": 1
}
",
        "sendBody": true,
        "sendHeaders": true,
        "specifyBody": "json",
        "headerParameters": {
          "parameters": [
            {
              "name": "Authorization",
              "value": "=Bearer {{ $json.perplexityAPI }}"
            }
          ]
        }
      },
      "typeVersion": 4.2
    },
    {
      "id": "c1ed633d-d318-403c-9577-c3c63ac2e68e",
      "name": "set API key",
      "type": "n8n-nodes-base.set",
      "position": [
        -120,
        0
      ],
      "parameters": {
        "options": {},
        "assignments": {
          "assignments": [
            {
              "id": "4f9bd3a0-5587-410f-b145-a287f65f9576",
              "name": "perplexityAPI",
              "type": "string",
              "value": "<yourPerplexityAPI>"
            }
          ]
        }
      },
      "typeVersion": 3.4
    },
    {
      "id": "e228e352-2ddd-4e2c-a434-993910ced7be",
      "name": "Post to X",
      "type": "n8n-nodes-base.twitter",
      "position": [
        320,
        0
      ],
      "parameters": {
        "text": "={{ $json.choices[0].message.content }}",
        "additionalFields": {}
      },
      "credentials": {
        "twitterOAuth2Api": {
          "id": "NY8wGzcN4f9f1UN4",
          "name": "X account 2 for images"
        }
      },
      "typeVersion": 2
    }
  ],
  "active": true,
  "pinData": {},
  "settings": {
    "callerPolicy": "workflowsFromSameOwner",
    "errorWorkflow": "qFOYM3IA9QZ7fMym",
    "executionOrder": "v1"
  },
  "versionId": "74e316fe-561f-4c80-b446-bba795654cef",
  "connections": {
    "Post to X": {
      "main": [
        []
      ]
    },
    "Perplexity": {
      "main": [
        [
          {
            "node": "Post to X",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "searchQuery": {
      "main": [
        [
          {
            "node": "set API key",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "set API key": {
      "main": [
        [
          {
            "node": "Perplexity",
            "type": "main",
            "index": 0
          }
        ]
      ]
    },
    "Schedule Trigger": {
      "main": [
        [
          {
            "node": "searchQuery",
            "type": "main",
            "index": 0
          }
        ]
      ]
    }
  }
}

功能特点

  • 自动检测新邮件
  • AI智能内容分析
  • 自定义分类规则
  • 批量处理能力
  • 详细的处理日志

技术分析

节点类型及作用

  • Scheduletrigger
  • Set
  • Httprequest
  • Twitter

复杂度评估

配置难度:
★★★☆☆
维护难度:
★★☆☆☆
扩展性:
★★★★☆

实施指南

前置条件

  • 有效的Gmail账户
  • n8n平台访问权限
  • Google API凭证
  • AI分类服务订阅

配置步骤

  1. 在n8n中导入工作流JSON文件
  2. 配置Gmail节点的认证信息
  3. 设置AI分类器的API密钥
  4. 自定义分类规则和标签映射
  5. 测试工作流执行
  6. 配置定时触发器(可选)

关键参数

参数名称 默认值 说明
maxEmails 50 单次处理的最大邮件数量
confidenceThreshold 0.8 分类置信度阈值
autoLabel true 是否自动添加标签

最佳实践

优化建议

  • 定期更新AI分类模型以提高准确性
  • 根据邮件量调整处理批次大小
  • 设置合理的分类置信度阈值
  • 定期清理过期的分类规则

安全注意事项

  • 妥善保管API密钥和认证信息
  • 限制工作流的访问权限
  • 定期审查处理日志
  • 启用双因素认证保护Gmail账户

性能优化

  • 使用增量处理减少重复工作
  • 缓存频繁访问的数据
  • 并行处理多个邮件分类任务
  • 监控系统资源使用情况

故障排除

常见问题

邮件未被正确分类

检查AI分类器的置信度阈值设置,适当降低阈值或更新训练数据。

Gmail认证失败

确认Google API凭证有效且具有正确的权限范围,重新进行OAuth授权。

调试技巧

  • 启用详细日志记录查看每个步骤的执行情况
  • 使用测试邮件验证分类逻辑
  • 检查网络连接和API服务状态
  • 逐步执行工作流定位问题节点

错误处理

工作流包含以下错误处理机制:

  • 网络超时自动重试(最多3次)
  • API错误记录和告警
  • 处理失败邮件的隔离机制
  • 异常情况下的回滚操作