AI-Powered Information Monitoring with OpenAI, Google Sheets, Jina AI and Slack
工作流概述
这是一个包含31个节点的复杂工作流，主要用于自动化处理各种任务。
工作流源代码
{
"id": "Xk0W98z9DVrNHeku",
"meta": {
"instanceId": "b9faf72fe0d7c3be94b3ebff0778790b50b135c336412d28fd4fca2cbbf8d1f5",
"templateCredsSetupCompleted": true
},
"name": "AI-Powered Information Monitoring with OpenAI, Google Sheets, Jina AI and Slack",
"tags": [],
"nodes": [
{
"id": "704de862-43e5-4322-ae35-45b505e68bb6",
"name": "OpenAI Chat Model",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
4220,
380
],
"parameters": {
"options": {}
},
"credentials": {
"openAiApi": {
"id": "",
"name": "OpenAi Connection"
}
},
"typeVersion": 1.1
},
{
"id": "eaae54b0-0500-47a7-ad8f-097e0882d21c",
"name": "Basic LLM Chain",
"type": "@n8n/n8n-nodes-langchain.chainLlm",
"position": [
4180,
-120
],
"parameters": {
"text": "={{ $json.data }}",
"messages": {
"messageValues": [
{
"message": "=You are an AI assistant responsible for summarizing articles **in English** and formatting them into Slack-compatible messages.
Your job is to create a clear and concise summary following the guidelines below and format it in Slack-specific Markdown format.
---
## 1. Title with Link
- Format the article title as a **clickable link** using Slack's Markdown syntax:
`<URL|*Title of the article*>`.
- The title should be clear and engaging to encourage readers to click.
---
## 2. Section Headings
- Use **bold text** to introduce different sections of the summary by wrapping the text with `*` symbols.
- Ensure headings are descriptive and guide the reader through the content effectively.
---
## 3. Key Points
- Present key insights using **bullet points**, using the `•` symbol for listing important information.
- Each point should be concise, informative, and directly related to the article's topic.
---
## 4. Content Summary
- Provide a brief but comprehensive overview of the article's content.
- Use plain text and line breaks to separate paragraphs for improved readability.
- Focus on the most important aspects without unnecessary details.
---
## 5. Context and Relevance
- Explain why the article is important and how it relates to the reader's interests.
- Highlight its relevance to ongoing trends or industry developments.
---
## Message Structure
The output should follow this structured format:
1. **Title with link** – Present the article as a clickable link formatted in Slack Markdown.
2. **Summary sections** – Organized under clear headings to enhance readability.
3. **Key insights** – Presented as bullet points for quick scanning.
4. **Contextual analysis** – A brief explanation of the article's relevance and importance.
---
## Slack Markdown Formatting Guide
Ensure the message follows Slack's Markdown syntax for proper display:
- **Bold text:** Use `*bold text*`.
- **Italic text:** Use `_italic text_`.
- **Bullet points:** Use `•` or `-` for lists.
- **Links:** Format as `<URL|*text*>` to create clickable links.
- **Line breaks:** Use a blank line to separate paragraphs for readability.
---
## Example of Slack-formatted Output
🔔 *New article from n8n Blog*
<https://blog.n8n.io/self-hosted-ai/|*Introducing the Self-hosted AI Starter Kit: Run AI locally for privacy-first solutions*>
*Summary of the article*
n8n has launched the Self-hosted AI Starter Kit, a Docker Compose template designed to simplify the deployment of local AI tools. This initiative addresses the growing need for on-premise AI solutions that enhance data privacy and reduce reliance on external APIs. The starter kit includes tools like Ollama, Qdrant, and PostgreSQL, providing a foundation for building self-hosted AI workflows. While it's tailored for proof-of-concept projects, users can customize it to fit specific requirements.
*Key Points*
• The Self-hosted AI Starter Kit facilitates quick setup of local AI environments using Docker Compose.
• It includes preconfigured AI workflow templates and essential tools such as Ollama, Qdrant, and PostgreSQL.
• Running AI on-premise offers benefits like improved data privacy and cost savings by minimizing dependence on external API calls.
• The kit is designed for easy deployment on local machines or personal cloud instances like Digital Ocean and runpod.io.
• n8n emphasizes the flexibility of their platform, allowing integration with over 400 services, including Google, Slack, Twilio, and JIRA, to streamline AI application development.
*Context and Relevance*
This article introduces a practical solution for organizations and developers seeking to implement AI workflows locally. By providing a ready-to-use starter kit, n8n addresses common challenges associated with setting up and maintaining on-premise AI systems, promoting greater control over data and potential cost efficiencies.
---
Ensure that the message is formatted according to Slack's requirements to improve readability and engagement.
"
}
]
},
"promptType": "define"
},
"typeVersion": 1.5
},
{
"id": "a3a10ccd-26f9-4b05-a79f-8754f619c153",
"name": "Schedule Trigger",
"type": "n8n-nodes-base.scheduleTrigger",
"position": [
-840,
120
],
"parameters": {
"rule": {
"interval": [
{
"field": "minutes",
"minutesInterval": 15
}
]
}
},
"typeVersion": 1.2
},
{
"id": "54ed8957-39be-4ad4-bea7-f56308d75a91",
"name": "RSS Read",
"type": "n8n-nodes-base.rssFeedRead",
"onError": "continueRegularOutput",
"position": [
800,
120
],
"parameters": {
"url": "={{ $json.rss_feed_url }}",
"options": {
"ignoreSSL": false
}
},
"executeOnce": false,
"typeVersion": 1.1
},
{
"id": "1ec53a9a-ca21-4da2-ab94-55b863a27aff",
"name": "Relevance Classification for Topic Monitoring",
"type": "@n8n/n8n-nodes-langchain.textClassifier",
"position": [
2380,
-20
],
"parameters": {
"options": {
"fallback": "discard"
},
"inputText": "={{ $json.title }}
{{ $json.contentSnippet }}",
"categories": {
"categories": [
{
"category": "relevant",
"description": "Articles related to artificial intelligence (AI), data science, machine learning, algorithms, big data, or innovations in these fields."
},
{
"category": "not_relevant",
"description": "Articles not directly related to artificial intelligence (AI), data science, machine learning, algorithms, big data, or innovations in these fields."
}
]
}
},
"typeVersion": 1
},
{
"id": "840431b1-cf2e-45e2-a79c-cab90f46a452",
"name": "Sticky Note",
"type": "n8n-nodes-base.stickyNote",
"position": [
2240,
-480
],
"parameters": {
"color": 7,
"width": 600,
"height": 960,
"content": "## LLM Call 1 - Article Topic Relevance Classification
This **LLM call** is used to **classify** whether the articles published on the website are **relevant** to the **topics and interests** you want to monitor.
It analyzes the **title** and the **content snippet** retrieved from the **RSS Read** node.
In this template, the monitored articles are related to **data and AI.**
The classification is done into **two categories**, which you should modify in the `Description` field under the **Categories** section of the node:
### Relevant
`Description`: Articles related to **[The topics you want to monitor]**.
### Not Relevant
`Description`: Articles that are not directly related to **[The topics you want to monitor]**.
By default, this template monitors topics related to artificial intelligence (AI), data science, machine learning, algorithms, big data, and innovations in these fields.
"
},
"typeVersion": 1
},
{
"id": "7dbc2246-9e1a-4c2e-a051-703e10e5fa0e",
"name": "Sticky Note3",
"type": "n8n-nodes-base.stickyNote",
"position": [
4020,
-660
],
"parameters": {
"color": 7,
"width": 600,
"height": 680,
"content": "## LLM Call 2 - Summarize and Format in Slack Markdown
This node **uses OpenAI's GPT-4o-mini model** to **summarize the article content**, which is provided as **Markdown text** from Jina AI, and formats it in **Slack Markdown** to enhance readability within Slack.
### Customize to fit your needs
Here are two examples of how you can modify the **System Prompt** of this node to better suit your requirements:
- **Language customization:**
You can modify the **System Prompt** to instruct the LLM to generate the summary in a specific language (e.g., French or Italian).
However, consider the option of adding a separate LLM node **dedicated to translation** if the model cannot handle **summarization, formatting, and translation** simultaneously while maintaining high output quality.
- **Changing the summary structure:**
You can adjust the prompt to modify how the summary is structured to better match your preferred format and style.
"
},
"typeVersion": 1
},
{
"id": "b472f924-81d9-4b99-8620-d95b286800c5",
"name": "Google Sheets - Get RSS Feed url followed",
"type": "n8n-nodes-base.googleSheets",
"position": [
260,
120
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": "gid=0",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit#gid=0",
"cachedResultName": "rss_feed"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit?usp=drivesdk",
"cachedResultName": "Template - AI-Powered Information Monitoring"
},
"authentication": "serviceAccount"
},
"credentials": {
"googleApi": {
"id": "",
"name": "Google Sheets account"
}
},
"executeOnce": true,
"typeVersion": 4.5
},
{
"id": "c2a571f0-614f-41cf-b0b0-db4c714a8ab8",
"name": "Sticky Note4",
"type": "n8n-nodes-base.stickyNote",
"position": [
80,
-480
],
"parameters": {
"color": 7,
"width": 460,
"height": 960,
"content": "## Google Sheets - Get RSS Feed URLs Followed
This node **retrieves rows** from the Google Sheet that contains the **RSS feed URLs** you follow.
It is configured to run only once per execution, meaning that even if the previous node outputs many items, this node will execute only once.
You can **add more URLs** to your sheet, but keep in mind that following **more RSS feeds** will increase the **cost of LLM API usage** (e.g., OpenAI).
You can access the **Google Sheet template** to copy and use in this workflow [here](https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/).
(*This is the same template used in the previous node.*)
In this node, make sure to select the **\"rss_feed\"** sheet from your **copied version of the Google Sheet template**.
This sheet contains the list of RSS feed URLs that the workflow will process."
},
"typeVersion": 1
},
{
"id": "90e34a2f-f326-4c83-ae26-d8f38d983c21",
"name": "Sticky Note6",
"type": "n8n-nodes-base.stickyNote",
"position": [
620,
-480
],
"parameters": {
"color": 7,
"width": 460,
"height": 960,
"content": "## RSS Read
This node **reads** the RSS feed.
The RSS URL is **retrieved** from the data you have entered in **Google Sheets**, so make sure the URL provided is indeed a **valid RSS feed**.
### What is an RSS feed?
An **RSS feed** is a **web feed** that allows users to **automatically receive updates** from websites, such as **news sites** or **blogs**, in a **standardized format**.
"
},
"typeVersion": 1
},
{
"id": "06c22fcc-6fb6-4646-8cd2-3e2c48a56fbc",
"name": "Sticky Note5",
"type": "n8n-nodes-base.stickyNote",
"position": [
2940,
-480
],
"parameters": {
"color": 7,
"width": 960,
"height": 500,
"content": "## Jina AI - Read URL
This node **uses the Jina AI API** to **retrieve the content** of articles that were classified as **\"relevant\"** in the previous step.
Since this process **involves web scraping**, ensure that it complies with the **scraping regulations** in your country.
### What is Jina AI?
**Jina AI** is an API that allows you to **extract webpage content** and convert it into a format that is **ready for LLM processing**, such as **Markdown**.
You can create an account [here](https://jina.ai/) and receive **1,000,000 free tokens** for testing.
However, the service can also be used **without an API key** (without an account), though with **reduced RPM (requests per minute)**.
For this workflow, the default RPM limits should generally be sufficient.
"
},
"typeVersion": 1
},
{
"id": "3f8a0ce3-d7b3-400b-bc03-1a233f441429",
"name": "Slack1",
"type": "n8n-nodes-base.slack",
"position": [
4940,
-120
],
"webhookId": "",
"parameters": {
"text": "={{ $json.text }}",
"select": "channel",
"channelId": {
"__rl": true,
"mode": "list",
"value": "C0898R9G7JP",
"cachedResultName": "topic-monitoring"
},
"otherOptions": {},
"authentication": "oAuth2"
},
"credentials": {
"slackOAuth2Api": {
"id": "",
"name": "slack-topic-monitoring"
}
},
"typeVersion": 2.3
},
{
"id": "6920300f-fd0e-41dc-adf6-ed5a3a267b3f",
"name": "Sticky Note8",
"type": "n8n-nodes-base.stickyNote",
"position": [
-460,
-480
],
"parameters": {
"color": 7,
"width": 460,
"height": 960,
"content": "## Google Sheets - Get Article Monitored Database
This node **retrieves rows** from the Google Sheet that contains articles **already monitored and summarized** by the workflow.
Depending on the RSS feed you monitor, **URLs may remain in the feed for a long time**, and you don't want to monitor the same URL **twice**.
You can find the **Google Sheet template** that you can copy and use in this workflow [here](https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit?gid=1966921272#gid=1966921272).
In this node, make sure to select the **\"article_database\"** sheet from your **copied version of the Google Sheet template**.
This sheet is used to store and manage the articles processed by the workflow.
---
## Set Field - existing_url
This node sets the **\"existing_url\"** field with the value from **\"article_url\"** in the Google Sheets database.
During the **first execution** of the workflow, this field will be **empty**, as no articles are present in Google Sheets yet.
An error may occur in this case; however, the workflow will **continue running** without interruption.
"
},
"typeVersion": 1
},
{
"id": "204aab36-1081-4d6e-b3a3-2fc03b6a1a10",
"name": "Sticky Note9",
"type": "n8n-nodes-base.stickyNote",
"position": [
1180,
-480
],
"parameters": {
"color": 7,
"width": 980,
"height": 960,
"content": "## Code Node to Filter Existing URLs
This code node filters URLs that have **not yet been summarized by AI.**
It outputs:
- A **list of URLs** following the RSS Read schema if new URLs are found.
- An item called **\"message\"** with the value **\"No new articles found\"** if no new articles are available in your RSS feed.
---
## IF Node
The condition for this node is: `{{ $json.message }}` *not equal to* **\"No new articles found\"**.
- **False** → The workflow executes the \"No Operation, do nothing\" node.
- **True** → The workflow proceeds to process the new articles for your web development industry monitoring.
"
},
"typeVersion": 1
},
{
"id": "ef83c5f9-12a7-4924-9356-d1307fc8f279",
"name": "Sticky Note10",
"type": "n8n-nodes-base.stickyNote",
"position": [
2940,
60
],
"parameters": {
"color": 7,
"width": 960,
"height": 580,
"content": "## Set Fields - Not Relevant Articles
This node prepares the data to be added to the Google Sheet by defining the following fields:
- **`article_url`** – The article's URL.
- **`summarized`** – Always set to `\"NO (not relevant)\"`, as it belongs to the **\"not_relevant\"** path.
- **`website`** – The website where the article URL was published.
- **`fetched_at`** – The timestamp when the URL was processed by the workflow.
> *(Note: This timestamp reflects when the scenario was triggered, as obtained from the **Schedule Trigger** node, not the exact fetch time.)*
- **`publish_date`** – The date the article was published.
---
## Google Sheets - Add Not Relevant Articles
This node adds the prepared data to the **\"article_database\"** sheet in your copied Google Sheet template.
Ensure that you select the **\"article_database\"** sheet when configuring this node.
"
},
"typeVersion": 1
},
{
"id": "10af053d-23f6-416b-9fe2-874dfc2ec7aa",
"name": "Sticky Note2",
"type": "n8n-nodes-base.stickyNote",
"position": [
4020,
80
],
"parameters": {
"color": 5,
"width": 600,
"height": 440,
"content": "## OpenAI Chat Model
This node specifies the **AI model** to be used for processing.
The default model is **GPT-4o-mini**, which has been **tested** and proven to perform well for this task.
**GPT-4o-mini** is a **cost-efficient** model, offering a good balance between **performance and affordability**, making it suitable for regular usage without incurring high costs.
"
},
"typeVersion": 1
},
{
"id": "67e6b0f9-32fc-4dcf-ae1b-effe11b31cd1",
"name": "Sticky Note11",
"type": "n8n-nodes-base.stickyNote",
"position": [
4680,
-640
],
"parameters": {
"color": 7,
"width": 600,
"height": 680,
"content": "## Slack - Send Article Summary
This node **posts the message** to the designated Slack channel, containing the **output generated by the LLM.**
For better organization and accessibility, it is recommended to use a **dedicated Slack channel** specifically for topic monitoring.
This ensures that team members can easily access relevant summaries without cluttering other discussions.
### Why not use Slack Tool Calling?
After extensive testing, the output from the previous node has proven to be **highly effective**, making it unnecessary to use **tool calling** or an **AI agent.** 😀
Keeping things simple **streamlines the workflow** and reduces complexity.
"
},
"typeVersion": 1
},
{
"id": "afe7643d-618b-4798-851e-b8b9d024e792",
"name": "Sticky Note12",
"type": "n8n-nodes-base.stickyNote",
"position": [
4700,
80
],
"parameters": {
"color": 7,
"width": 1260,
"height": 560,
"content": "## Set Fields - Relevant Articles
This node prepares the data to be added to the Google Sheet by defining the following fields:
- **`article_url`** – The article's URL.
- **`summarized`** – Always set to `\"YES\"`, as it follows the **\"relevant\"** path.
- **`summary`** – The article summary that was posted to Slack.
- **`website`** – The source website where the article was published.
- **`fetched_at`** – The timestamp indicating when the URL was processed by the workflow.
> *(Note: This timestamp reflects when the data was added to Google Sheets, not the actual fetch time.)*
- **`publish_date`** – The date the article was published.
---
## Google Sheets - Add Relevant Articles
This node adds the prepared data to the **\"article_database\"** sheet in your copied Google Sheet template.
Make sure to select the **\"article_database\"** sheet when configuring this node.
"
},
"typeVersion": 1
},
{
"id": "e87619df-48e3-4ef8-83c7-1695746e2b92",
"name": "Sticky Note1",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1000,
-280
],
"parameters": {
"color": 7,
"width": 460,
"height": 600,
"content": "## Scheduler
This **trigger** is a **scheduler** that defines **how often the workflow is executed**.
By default, the **template is set to every 1 hour**, meaning the workflow will check **every hour** if **new articles** have been added to the **RSS feed** you follow.
"
},
"typeVersion": 1
},
{
"id": "e2bcd684-abd9-4f47-bf4c-12eac379432d",
"name": "Sticky Note7",
"type": "n8n-nodes-base.stickyNote",
"position": [
-1900,
-720
],
"parameters": {
"color": 6,
"width": 780,
"height": 1300,
"content": "# Workflow Overview
## Check Legal Regulations:
This workflow involves scraping, so ensure you comply with the legal regulations in your country before getting started. Better safe than sorry!
## 📌 Purpose
This workflow enables **automated and AI-driven topic monitoring**, delivering **concise article summaries** directly to a **Slack channel** in a structured and easy-to-read format.
It allows users to stay informed on specific topics of interest effortlessly, without manually checking multiple sources, ensuring a **time-efficient and focused** monitoring experience.
**To get started, copy the Google Sheets template required for this workflow from [here](https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY).**
## 🎯 Target Audience
This workflow is designed for:
- **Industry professionals** looking to track key developments in their field.
- **Research teams** who need up-to-date insights on specific topics.
- **Companies** aiming to keep their teams informed with relevant content.
## ⚙️ How It Works
1. **Trigger:** A **Scheduler** initiates the workflow at regular intervals (default: every hour).
2. **Data Retrieval:**
- RSS feeds are fetched using the **RSS Read** node.
- Previously monitored articles are checked in **Google Sheets** to avoid duplicates.
3. **Content Processing:**
- The article relevance is assessed using **OpenAI (GPT-4o-mini)**.
- Relevant articles are scraped using **Jina AI** to extract content.
- Summaries are generated and formatted for Slack.
4. **Output:**
- Summaries are posted to the specified Slack channel.
- Article metadata is stored in **Google Sheets** for tracking.
## 🛠️ Key APIs and Nodes Used
- **Scheduler Node:** Triggers the workflow periodically.
- **RSS Read:** Fetches the latest articles from defined RSS feeds.
- **Google Sheets:** Stores monitored articles and manages feed URLs.
- **OpenAI API (GPT-4o-mini):** Classifies article relevance and generates summaries.
- **Jina AI API:** Extracts the full content of relevant articles.
- **Slack API:** Posts formatted messages to Slack channels.
---
This workflow provides an **efficient and intelligent way** to stay informed about your topics of interest, directly within Slack.
"
},
"typeVersion": 1
},
{
"id": "d72f505d-2bbf-41db-b404-8a61b8c21452",
"name": "Google Sheets - Get article monitored database",
"type": "n8n-nodes-base.googleSheets",
"position": [
-400,
120
],
"parameters": {
"options": {},
"sheetName": {
"__rl": true,
"mode": "list",
"value": 1966921272,
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit#gid=1966921272",
"cachedResultName": "article_database"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit?usp=drivesdk",
"cachedResultName": "Template - AI-Powered Information Monitoring"
},
"authentication": "serviceAccount"
},
"credentials": {
"googleApi": {
"id": "",
"name": "Google Sheets account"
}
},
"executeOnce": true,
"typeVersion": 4.5,
"alwaysOutputData": true
},
{
"id": "08eae799-2682-4d49-81fa-2127a65d887b",
"name": "Code",
"type": "n8n-nodes-base.code",
"position": [
1280,
120
],
"parameters": {
"jsCode": "// Retrieve data from RSS feed and Google Sheets
const rssItems = items; // Contains RSS articles
const sheetItems = $items(\"Set field - existing_url\", 0);
// Extract the links of articles present in Google Sheets
const existingUrls = sheetItems.map(entry => entry.json.existing_url);
// Filter RSS articles to keep only those not present in Google Sheets
const newArticles = rssItems.filter(rssItem => {
return !existingUrls.includes(rssItem.json.link);
});
// If new articles are found, return them
if (newArticles.length > 0) {
return newArticles;
}
// If no new articles, return an informational message
return [{ json: { message: \"No new articles found.\" } }];
"
},
"typeVersion": 2
},
{
"id": "9f2d2c87-460b-4872-9538-519d26524475",
"name": "No Operation, do nothing",
"type": "n8n-nodes-base.noOp",
"position": [
1960,
240
],
"parameters": {},
"typeVersion": 1
},
{
"id": "e9ebbce6-a3b4-4f89-9908-3d9b2dd42f44",
"name": "If",
"type": "n8n-nodes-base.if",
"position": [
1640,
120
],
"parameters": {
"options": {},
"conditions": {
"options": {
"version": 2,
"leftValue": "",
"caseSensitive": true,
"typeValidation": "strict"
},
"combinator": "and",
"conditions": [
{
"id": "bad6fc33-2e1e-4169-9893-d284c6c68288",
"operator": {
"type": "string",
"operation": "notEquals"
},
"leftValue": "={{ $json.message }}",
"rightValue": "No new articles found."
}
]
}
},
"typeVersion": 2.2
},
{
"id": "6e2c820d-27da-4d3b-844c-581fb266e04a",
"name": "Jina AI - Read URL",
"type": "n8n-nodes-base.httpRequest",
"position": [
3240,
-120
],
"parameters": {
"url": "=https://r.jina.ai/{{ $json.link }}",
"options": {}
},
"retryOnFail": true,
"typeVersion": 4.2,
"waitBetweenTries": 5000
},
{
"id": "3f942518-f75b-4d03-9cd1-b275ad3b91cd",
"name": "Set field - existing_url",
"type": "n8n-nodes-base.set",
"onError": "continueRegularOutput",
"position": [
-180,
120
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "07799638-55d7-42a9-b1f7-fea762cfa2f1",
"name": "existing_url",
"type": "string",
"value": "={{ $json.article_url.extractUrl() }}"
}
]
}
},
"typeVersion": 3.4,
"alwaysOutputData": true
},
{
"id": "baef0ff9-8bf5-4ecf-9300-0adbad0d1a07",
"name": "OpenAI Chat Model1",
"type": "@n8n/n8n-nodes-langchain.lmChatOpenAi",
"position": [
2400,
300
],
"parameters": {
"options": {}
},
"credentials": {
"openAiApi": {
"id": "",
"name": "OpenAi Connection"
}
},
"typeVersion": 1.1
},
{
"id": "ccbfe5fc-2e87-4fff-b23d-0c4c6ebd3648",
"name": "Set fields - Not relevant articles",
"type": "n8n-nodes-base.set",
"position": [
3060,
480
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "3fbf5256-f06b-450a-adf7-65591a19c7dd",
"name": "article_url",
"type": "string",
"value": "={{ $json.link }}"
},
{
"id": "02f506cf-28fe-46ef-b97e-7ec938805151",
"name": "summarized",
"type": "string",
"value": "NO (not relevant)"
},
{
"id": "552efef4-63cb-448b-bb0c-30ae9666f310",
"name": "website",
"type": "string",
"value": "={{ $('Google Sheets - Get RSS Feed url followed').item.json.website }}"
},
{
"id": "096acb35-4e9e-48fd-8e61-8ceb525591fa",
"name": "fetched_at",
"type": "string",
"value": "={{$now}}"
},
{
"id": "427243d1-01c4-458a-9626-75366e4264cd",
"name": "publish_date",
"type": "string",
"value": "={{ $('Relevance Classification for Topic Monitoring').item.json.pubDate.toDateTime().format('yyyy-MM-dd') }}"
}
]
}
},
"typeVersion": 3.4
},
{
"id": "0dbcc872-9afa-4e2c-be24-82d3a2457dd0",
"name": "Google Sheets - Add relevant articles",
"type": "n8n-nodes-base.googleSheets",
"position": [
3480,
480
],
"parameters": {
"columns": {
"value": {},
"schema": [
{
"id": "article_url",
"type": "string",
"display": true,
"required": false,
"displayName": "article_url",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "summarized",
"type": "string",
"display": true,
"required": false,
"displayName": "summarized",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "summary",
"type": "string",
"display": true,
"required": false,
"displayName": "summary",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "website",
"type": "string",
"display": true,
"required": false,
"displayName": "website",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "fetched_at",
"type": "string",
"display": true,
"required": false,
"displayName": "fetched_at",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "publish_date",
"type": "string",
"display": true,
"required": false,
"displayName": "publish_date",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "autoMapInputData",
"matchingColumns": [],
"attemptToConvertTypes": false,
"convertFieldsToString": false
},
"options": {},
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "list",
"value": 1966921272,
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit#gid=1966921272",
"cachedResultName": "article_database"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit?usp=drivesdk",
"cachedResultName": "Template - AI-Powered Information Monitoring"
},
"authentication": "serviceAccount"
},
"credentials": {
"googleApi": {
"id": "",
"name": "Google Sheets account"
}
},
"typeVersion": 4.5
},
{
"id": "0c7024b6-dfac-4e97-9d42-198fff6bcc47",
"name": "Google Sheets - Add relevant article",
"type": "n8n-nodes-base.googleSheets",
"position": [
5660,
520
],
"parameters": {
"columns": {
"value": {},
"schema": [
{
"id": "article_url",
"type": "string",
"display": true,
"required": false,
"displayName": "article_url",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "summarized",
"type": "string",
"display": true,
"required": false,
"displayName": "summarized",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "summary",
"type": "string",
"display": true,
"required": false,
"displayName": "summary",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "website",
"type": "string",
"display": true,
"required": false,
"displayName": "website",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "fetched_at",
"type": "string",
"display": true,
"required": false,
"displayName": "fetched_at",
"defaultMatch": false,
"canBeUsedToMatch": true
},
{
"id": "publish_date",
"type": "string",
"display": true,
"required": false,
"displayName": "publish_date",
"defaultMatch": false,
"canBeUsedToMatch": true
}
],
"mappingMode": "autoMapInputData",
"matchingColumns": [],
"attemptToConvertTypes": false,
"convertFieldsToString": false
},
"options": {},
"operation": "append",
"sheetName": {
"__rl": true,
"mode": "list",
"value": 1966921272,
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit#gid=1966921272",
"cachedResultName": "article_database"
},
"documentId": {
"__rl": true,
"mode": "list",
"value": "1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY",
"cachedResultUrl": "https://docs.google.com/spreadsheets/d/1F2FzWt9FMkA5V5i9d_hBJRahLDvxs3DQBOLkLYowXbY/edit?usp=drivesdk",
"cachedResultName": "Template - AI-Powered Information Monitoring"
},
"authentication": "serviceAccount"
},
"credentials": {
"googleApi": {
"id": "",
"name": "Google Sheets account"
}
},
"typeVersion": 4.5
},
{
"id": "e1266606-eaee-4077-be7e-6f08ae9bae39",
"name": "Set Fields - Relevant Articles",
"type": "n8n-nodes-base.set",
"position": [
4900,
520
],
"parameters": {
"options": {},
"assignments": {
"assignments": [
{
"id": "3fbf5256-f06b-450a-adf7-65591a19c7dd",
"name": "article_url",
"type": "string",
"value": "={{ $('Relevance Classification for Topic Monitoring').item.json.link }}"
},
{
"id": "02f506cf-28fe-46ef-b97e-7ec938805151",
"name": "summarized",
"type": "string",
"value": "YES"
},
{
"id": "e23059bd-8bb2-439a-85bd-f9e191930d1e",
"name": "summary",
"type": "string",
"value": "={{ $json.text }}"
},
{
"id": "552efef4-63cb-448b-bb0c-30ae9666f310",
"name": "website",
"type": "string",
"value": "={{ $('Google Sheets - Get RSS Feed url followed').item.json.website }}"
},
{
"id": "096acb35-4e9e-48fd-8e61-8ceb525591fa",
"name": "fetched_at",
"type": "string",
"value": "={{$now}}"
},
{
"id": "427243d1-01c4-458a-9626-75366e4264cd",
"name": "publish_date",
"type": "string",
"value": "={{ $('Relevance Classification for Topic Monitoring').item.json.pubDate.toDateTime().format('yyyy-MM-dd') }}"
}
]
}
},
"typeVersion": 3.4
}
],
"active": false,
"pinData": {},
"settings": {
"executionOrder": "v1"
},
"versionId": "dcc84e7c-aa42-4d0f-8522-84fdf8bea0bc",
"connections": {
"If": {
"main": [
[
{
"node": "Relevance Classification for Topic Monitoring",
"type": "main",
"index": 0
}
],
[
{
"node": "No Operation, do nothing",
"type": "main",
"index": 0
}
]
]
},
"Code": {
"main": [
[
{
"node": "If",
"type": "main",
"index": 0
}
]
]
},
"RSS Read": {
"main": [
[
{
"node": "Code",
"type": "main",
"index": 0
}
]
]
},
"Basic LLM Chain": {
"main": [
[
{
"node": "Slack1",
"type": "main",
"index": 0
},
{
"node": "Set Fields - Relevant Articles",
"type": "main",
"index": 0
}
]
]
},
"Schedule Trigger": {
"main": [
[
{
"node": "Google Sheets - Get article monitored database",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model": {
"ai_languageModel": [
[
{
"node": "Basic LLM Chain",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Jina AI - Read URL": {
"main": [
[
{
"node": "Basic LLM Chain",
"type": "main",
"index": 0
}
]
]
},
"OpenAI Chat Model1": {
"ai_languageModel": [
[
{
"node": "Relevance Classification for Topic Monitoring",
"type": "ai_languageModel",
"index": 0
}
]
]
},
"Set field - existing_url": {
"main": [
[
{
"node": "Google Sheets - Get RSS Feed url followed",
"type": "main",
"index": 0
}
]
]
},
"Set Fields - Relevant Articles": {
"main": [
[
{
"node": "Google Sheets - Add relevant article",
"type": "main",
"index": 0
}
]
]
},
"Set fields - Not relevant articles": {
"main": [
[
{
"node": "Google Sheets - Add relevant articles",
"type": "main",
"index": 0
}
]
]
},
"Google Sheets - Add relevant article": {
"main": [
[]
]
},
"Google Sheets - Get RSS Feed url followed": {
"main": [
[
{
"node": "RSS Read",
"type": "main",
"index": 0
}
]
]
},
"Relevance Classification for Topic Monitoring": {
"main": [
[
{
"node": "Jina AI - Read URL",
"type": "main",
"index": 0
}
],
[
{
"node": "Set fields - Not relevant articles",
"type": "main",
"index": 0
}
]
]
},
"Google Sheets - Get article monitored database": {
"main": [
[
{
"node": "Set field - existing_url",
"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
- @N8N/N8N Nodes Langchain.Lmchatopenai
- @N8N/N8N Nodes Langchain.Chainllm
- Scheduletrigger
- Rssfeedread
- @N8N/N8N Nodes Langchain.Textclassifier
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
- 在n8n中导入工作流JSON文件
- 配置Gmail节点的认证信息
- 设置AI分类器的API密钥
- 自定义分类规则和标签映射
- 测试工作流执行
- 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