一条n8n工作流
一条n8n工作流
- 1 工作流环境信息
- 2 工作流编辑
- 3 工作流执行
- 4 工作流各模块
- 5 工作流json文件
这里是我的一条工作流，功能是：周期性地获取最新的热点信息，发送到我的钉钉机器人。文末附有完整的工作流json文件，以供有兴趣者参考。
近期使用n8n的主要心得就是：清爽、自由度高。它里面的大模型可以使用ollama、deepseek、openai等，使用开源的话，直接ollama中的7b或者1.5b参数的自己玩也是可以的。如果水平再高些，其实国外有很多免费使用的大模型API，比如groq家的70b参数的，但需要解决网络问题。
1 工作流环境信息
| 序号序号 | 工具 | 版本 | 大小 | 镜像 | 数据卷 |
|---|---|---|---|---|---|
| 1 | n8n | 1.104.2 | 1.02GB | docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/n8nio/n8n:1.104.2 | -v E:\n8n_data:/home/node/.n8n |
| 2 | postgres | 16.4 | 335MB | docker pull swr.cn-north-4.myhuaweicloud.com/ddn-k8s/docker.io/bitnami/postgresql:16 | -v E:\postgresql16:/bitnami/postgresql |
2 工作流编辑
工作流主要涉及AI Agent节点、数据库节点、code自定义节点，是简单工作流的完整呈现。
3 工作流执行
触发执行这里可以人工点、周期调度两种方式，目前周期调度正常，可以加载文末的json文件，然后修改主要参数就可以执行。
4 工作流各模块
这里的HTTP-Request节点，这里的热点信息使用的一个免费接口，添加即用。
https://api.1314.cool/getbaiduhot/
为了保存每天的热点历史信息，我这里使用了postgres数据库，数据库表信息，如下：
CREATE TABLE baidu_hot_words_history (id SERIAL PRIMARY KEY,hot_words VARCHAR(255) NOT NULL,hot_word_url TEXT,record_time TIMESTAMP WITH TIME ZONE DEFAULT CURRENT_TIMESTAMP,record_day DATE DEFAULT CURRENT_DATE
);
-- 添加表注释
COMMENT ON TABLE baidu_hot_words_history IS '百度热搜历史信息存储表';
-- 添加字段注释
COMMENT ON COLUMN baidu_hot_words_history.hot_words IS '热词（热搜关键词）';
COMMENT ON COLUMN baidu_hot_words_history.hot_word_url IS '热词链接（百度搜索URL）';
通过HTTP节点获取了热搜后，与当天已推送的历史信息比对，保证只推送新增的信息，我这里使用的python代码，当然也可以使用javascript。
if len(_('Select-获取历史数据').all()) == 0:return _('HTTP-百度热点').all()
else:new_hots = {}# http-requestvalue = []for items in _('HTTP-百度热点').all()[0]['json']['data']:# postgresnum = 0for select in _('Select-获取历史数据').all():if items['word'] == select['json']['hot_words']:# 历史信息里存在num += 1if num == 0:value.append(items)new_hots["data"] = valuereturn new_hots
这里添加了一个过滤判断，有新增才推送下一个节点。
获取的新增热点，有AI Agent进行整理后，推送给钉钉节点，我这里的prompt如下：
你是一位热搜信息整理专家。
任务：只整理获取的热搜信息{{$node["PyCode-数据比对"].json.toJsonString()}}，不产生推荐信息。
请严格按以下格式输出：
热点信息（多少条）:
1.热点信息1
2.热点信息2
我这里的chat模型使用的openai的，开源的话可以使用你本地的ollama的，如果你是RMB玩家，可以使用deepseek等，或者国外一些免费api，但是注意数据安全！！！
最后节点之一是钉钉接口，如何你还不太熟悉，可以参考我的这篇文章:调用钉钉接口实现机器人推送消息_钉钉机器人调用第三方接口-CSDN博客 。
最后节点之一是及时更新新增热点到历史数据中，也就是postgres中。
INSERT INTO baidu_hot_words_history (hot_words, hot_word_url)
VALUES
{{ $json.data.map(item => "('" + item.word.replace(/'/g, "''") + "','" + item.url.replace(/'/g, "''") + "')").join(",")
}};
5 工作流json文件
保存下面的文本为json文件，在n8n工作流编辑器页面点击从文件导入即可。
{"name": "Workflow_HotWord_To_Dingding_show","nodes": [{"parameters": {"url": "https://api.1314.cool/getbaiduhot/","options": {}},"type": "n8n-nodes-base.httpRequest","typeVersion": 4.2,"position": [-1088,96],"id": "945ea412-b1d8-4436-84e7-c24434928b68","name": "HTTP-百度热点"},{"parameters": {"operation": "select","schema": "public","table": "baidu_hot_words_history","returnAll": true,"where": {"values": [{"column": "record_day","value": "={{ new Date().toISOString().split('T')[0] }}"}]},"options": {}},"type": "n8n-nodes-base.postgres","typeVersion": 2.6,"position": [-864,96],"id": "83d30cd2-9cf7-4185-aabc-c32ee2110312","name": "Select-获取历史数据","credentials": {"postgres": {"id": "HXP7gxgwbVI8mlKX","name": "Postgres account"}}},{"parameters": {"language": "python","pythonCode": "if len(_('Select-获取历史数据').all()) == 0:\n return _('HTTP-百度热点').all()\nelse:\n new_hots = {}\n # http-request\n value = []\n for items in _('HTTP-百度热点').all()[0]['json']['data']:\n # postgres\n num = 0\n for select in _('Select-获取历史数据').all():\n if items['word'] == select['json']['hot_words']:\n # 历史信息里存在\n num += 1\n if num == 0:\n value.append(items)\n new_hots[\"data\"] = value\n return new_hots"},"type": "n8n-nodes-base.code","typeVersion": 2,"position": [-640,96],"id": "f00feedc-bd3a-4897-85dd-50acdd2e6279","name": "PyCode-数据比对","alwaysOutputData": false},{"parameters": {"conditions": {"options": {"caseSensitive": true,"leftValue": "","typeValidation": "strict","version": 2},"conditions": [{"id": "aaf8b4d1-f776-4626-82ba-b5c12783ba96","leftValue": "={{ $node[\"PyCode-数据比对\"].json.values()[0].length}}","rightValue": 0,"operator": {"type": "number","operation": "gt","name": "filter.operator.gt"}}],"combinator": "and"},"options": {}},"type": "n8n-nodes-base.if","typeVersion": 2.2,"position": [-416,96],"id": "b938433e-bbd5-433d-84cc-cb6be46aa60a","name": "If-新增热词","executeOnce": true},{"parameters": {"promptType": "define","text": "整理热点信息","options": {"systemMessage": "=你是一位热搜信息整理专家。\n任务：只整理获取的热搜信息{{$node[\"PyCode-数据比对\"].json.toJsonString()}}，不产生推荐信息。\n请严格按以下格式输出：\n热点信息（多少条）:\n1.热点信息1\n2.热点信息2\n\n\n\n\n"}},"type": "@n8n/n8n-nodes-langchain.agent","typeVersion": 2.1,"position": [-192,-160],"id": "036b0f07-0150-4068-96fa-cda634542c0c","name": "AI Agent","retryOnFail": false,"maxTries": 2,"executeOnce": true},{"parameters": {"method": "POST","url": "https://oapi.dingtalk.com/robot/send?access_token=你自己的token","sendHeaders": true,"headerParameters": {"parameters": [{"name": "Content-Type","value": "application/json"}]},"sendBody": true,"specifyBody": "json","jsonBody": "={\n \"msgtype\": \"text\",\n \"text\": {\n \"content\": {{$json.output.toJsonString()}}\n }\n}","options": {}},"type": "n8n-nodes-base.httpRequest","typeVersion": 4.2,"position": [208,-64],"id": "8102b92e-c490-4e39-bbbc-a262dc7de2a3","name": "钉钉通知"},{"parameters": {"operation": "executeQuery","query": "INSERT INTO baidu_hot_words_history (hot_words, hot_word_url)\nVALUES \n{{ \n $json.data.map(item => \n \"('\" + item.word.replace(/'/g, \"''\") + \"','\" + item.url.replace(/'/g, \"''\") + \"')\"\n ).join(\",\") \n}};","options": {}},"type": "n8n-nodes-base.postgres","typeVersion": 2.6,"position": [-112,240],"id": "c4f56f3a-4ca4-40bd-a32a-ea6b1e6caa53","name": "Insert-更新历史数据","credentials": {"postgres": {"id": "HXP7gxgwbVI8mlKX","name": "Postgres account"}}},{"parameters": {},"type": "n8n-nodes-base.manualTrigger","typeVersion": 1,"position": [-1312,0],"id": "fa3481d8-7f2d-4569-a4ab-d052ec57622b","name": "手动执行"},{"parameters": {"rule": {"interval": [{"field": "minutes"}]}},"type": "n8n-nodes-base.scheduleTrigger","typeVersion": 1.2,"position": [-1312,192],"id": "22a33d38-0fdd-4442-94c6-7340ce144ca6","name": "周期调度"},{"parameters": {"model": {"__rl": true,"mode": "list","value": "gpt-4.1-mini"},"options": {}},"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi","typeVersion": 1.2,"position": [-176,16],"id": "44b4a19d-d9b8-4b04-be56-74c254cdda76","name": "OpenAI Chat Model","credentials": {"openAiApi": {"id": "Hm8gyQkgPUkOVBi6","name": "OpenAi account"}}}],"pinData": {},"connections": {"HTTP-百度热点": {"main": [[{"node": "Select-获取历史数据","type": "main","index": 0}]]},"Select-获取历史数据": {"main": [[{"node": "PyCode-数据比对","type": "main","index": 0}]]},"PyCode-数据比对": {"main": [[{"node": "If-新增热词","type": "main","index": 0}]]},"If-新增热词": {"main": [[{"node": "AI Agent","type": "main","index": 0},{"node": "Insert-更新历史数据","type": "main","index": 0}],[]]},"AI Agent": {"main": [[{"node": "钉钉通知","type": "main","index": 0}]]},"手动执行": {"main": [[{"node": "HTTP-百度热点","type": "main","index": 0}]]},"周期调度": {"main": [[{"node": "HTTP-百度热点","type": "main","index": 0}]]},"钉钉通知": {"main": [[]]},"OpenAI Chat Model": {"ai_languageModel": [[{"node": "AI Agent","type": "ai_languageModel","index": 0}]]}},"active": false,"settings": {"executionOrder": "v1"},"versionId": "55bd8cd7-134f-4cc4-8a93-2cce358d5d94","meta": {"templateCredsSetupCompleted": true,"instanceId": "1f16bc26aa41d68046a6cd33330705e84dc6e662ed18adf783908c69e188965b"},"id": "0OYbYUyfETIADQVv","tags": []
}