How to repurpose Google news into blog post using Dumpling AI, Claude AI, in Make.com

This tutorial walks you through automating the process of transforming Google News articles into professional blog posts using Dumpling AI, Claude AI and Make.com. The scenario will search Google News for relevant articles, scrape their content, rewrite them into engaging blog posts using AI, generate metadata for SEO, and store everything neatly in Google Sheets. By the end of this tutorial, you’ll have a fully automated content creation pipeline for your blog, saving you hours of manual work.

Step 1: Set Up Your Make.com Scenario

1. Log in to Make.com and navigate to your dashboard.

Click on the “Create a new scenario” button. This will take you to the scenario setup interface, where you’ll build your automation from scratch.

Step 2: Adding the Dumpling AI News Search Module

Why use Dumpling AI for Google news search module?
The Dumpling AI module provides access to real-time news, and it allows you to search for specific topics and fetch articles from trusted news sources. This makes it a valuable tool for automating content repurposing based on current events or trends. The accuracy, speed, and filtering options in the module make it perfect for repurposing content into blog posts.

How to add and configure:

1. Click the plus (+) icon on the blank canvas and search for the Dumpling AI: Search News module.
2. Connect your Dumpling AI account by selecting your saved connection or adding a new one under the __IMTCONN__ field.
   • If you’re using Dumpling AI for the first time, you’ll need to authenticate your account in Make.com using API keys

Configure the search parameters:

• Query: Enter a relevant search term, such as AI Automation. This field is crucial for guiding the AI to find the most relevant news articles based on your needs.
• Limit: Set the limit to “1” optional. This defines the number of articles returned. For automation purposes, we keep this number low to focus on one article per run.

You can also add optional filters such as:

• Date Range: You can choose to search for articles from the past day, week, or even the past year, depending on how timely you want your content.
• Language: Set the language if you want news articles in a specific language (e.g., English).

Step 3: Scrape the Article Content with Dumpling AI

Why scrape the article?
Once you have the article link, you need the content to convert it into a blog post. Scraping the content ensures that the text from the article is extracted and formatted properly for further processing. The Dumpling AI scraper module helps you scrape this data from the article link in a clean and readable format.

1. Add the Dumpling AI: Scrape module next.
   • This module allows you to extract the text content from the URL provided by the previous step (news search).

The scraped content is cleaned up and converted into a readable format, such as Markdown, making it ideal for repurposing into blog posts.

1. Configure the scraper:
   • In the URL field, map the output from the previous step (the article link returned by the search module). Use {{24.link}} to reference the link dynamically.
   • Set the format to Markdown so the content is structured and easy to convert into a blog post.
   • Turn on the “Clean Data” option, which removes unwanted elements like ads or navigation content.

Step 4: Rewrite the Article as a Blog Post Using AI

1. Add the Anthropic Claude: Create a Message module.
   • This module allows you to use an AI model to rewrite the scraped content into a professional blog post.
2. Configure the module:
   • Model: Select clause-3-5-sonnet-20240620 (or another available model) for high-quality natural language generation.
   • Max Tokens: Set the limit to 1000 tokens to ensure the generated text covers the entire article.
   • Messages:
     • Role: Set this to “User”.
     • Content: The message here is the prompt for the AI model. Use {{25.content}} (the scraped content) to feed the article text into the AI for rewriting.
     • Provide clear instructions in the prompt for rewriting, such as: Please rewrite this article as a blog post while keeping the main points intact. Ensure the tone is professional but conversational, and break it into clear sections with subheadings.

Step 5: Generate Blog Metadata Using Claude AI 

1. Add another Claude AI: Create a Message module to generate structured metadata for your blog post.

1. Configure the module:
   • Model: Again, choose claude-3-5-sonnet-20240620.
   • Messages:
     • Role: “User.”
     • Content: Ask the AI to generate metadata in JSON format using the blog post from the previous step, here is the prompt used:

I have a blog post below. Create the following metadata for it in JSON format:  

1. **Title**: A clear, engaging title for the post.  

2. **Meta Description**: A brief, compelling summary of the blog post (under 160 characters).  

3. **Slug**: A short, SEO-friendly URL slug based on the title.  

4. **Root URL**: The full URL, using the slug appended to ‘https://myblog.com/’.  

5. **Category**: The most relevant blog category based on the content.  

Output everything in the following JSON structure:  

{

  “title”: “Generated Blog Post Title”,

  “meta_description”: “A brief summary of the blog post.”,

  “slug”: “generated-blog-post-title”,

  “root_url”: “https://myblog.com/generated-blog-post-title”,

  “category”: “Blog Category”

}

Step 5: Generate an Image for the Blog Post (Optional)

Using Dumpling AI, you can generate an image tailored to the content of the article.

1. Add the Dumpling AI: Generate Image module.
2. Configure the module:
   • Prompt: I have a blog post below. Create a visually appealing image that captures the essence of the blog post. The image should reflect the main themes and ideas discussed in the post and be suitable for sharing on social media. Make sure the image is engaging and relevant to the content. Here is the blog post  [{{26.textResponse}}].

Image Size: Set this to landscape_4_3 for a blog-friendly aspect ratio

Step 6: Parse the AI Output and Extract Metadata

Why parse the response?
The parse JSON module helps extract structured data, such as metadata for SEO, e.tc that was generated using the Claude AI Module. Here is how it is structured:

Add a JSON Parse module.

1. Configure the module:
   • In the JSON field, map the output from the previous AI-generated text.
   • This will extract data such as the blog title, meta description, and URL slug.

Step 7: Automatically Add the Blog Post to Google Sheets

1. Add a Google Sheets: Add Row module.
2. Configure the module:
   • Choose the spreadsheet and sheet where you want to store the blog post data.
   • Map fields such as:
     • Article URL: {{25.url}}
     • Meta Description: {{29.meta_description}}
     • Blog Title: {{29.title}}
     • Blog Post Content: {{26.textResponse}}
     • Image URL (if generated) from the Dumpling AI module.

Step 8: Test and Run the Scenario

1. Run the scenario once to test it and see if everything is working as expected.

After successful testing, schedule the scenario to run automatically at your desired intervals (daily, weekly, etc.) or set a trigger to run it whenever a new article is needed.

By following these steps, you can automate the entire process of transforming current news articles into polished blog posts, with images, and store them in Google Sheets for easy access.

Get the Blueprint Featured in This Guide

Access the full blueprint here to get started on setting up this automation effortlessly!