This flow will generate structured news articles on trending topics by retrieving top articles from Google. Using the “Inverted Pyramid” structure ensures that the most important information appears first, followed by supporting details, making it perfect for creating concise, impactful news articles.
All you need to do is ask about a topic. For the best results, input topics that are currently trending or receiving news coverage. The template will then search Google for news articles no older than 48 hours, passing the information to an AI Agent:
Try this Flow:
Once the AI Agent receives the information, it creates a structured article. It knows exactly what to do thanks to its System message:
“Your task is to generate latest news article about topic defined by human input. Enhance the human input query with words like ‘latest news’ or ‘last 24 hours’ or ‘today’ to make sure you find the latest news only.
Search top 5 urls ranking for latest news about given topic on Google Search.
Use URL Retriever tool to extract text from found urls.
Make sure news are latest from last 48 hours. Ignore older news.
News articles should be written in a structure known as the “inverted pyramid.” In the inverted pyramid format, the most newsworthy information goes at the beginning of the story and the least newsworthy information goes at the end.”
Components breakdown
Tip: Input specific, trending topics for more timely and relevant news articles.
Chat Input: This is the message you send in the chat. It’s the starting point of any flow.
URL Retriever: Allows your Flow to retrieve information from individual URLs.
GoogleSearch: Searches the web for results relevant to the user query.
Chat History: Ensures the bot remembers a set amount of previous messages.
Tool Calling Agent: A link to the AI. Agents are autonomous and use their own reasoning to perform tasks.
Chat Output: Component representing the chatbot’s answer.
Useful tips
There are several ways you can adjust and improve this flow:
Change the System message to reflect your exact needs.
Use an LLM component to switch to a different LLM model.
