Automate Web Research and Summarize Multiple Sources

Turn any search query into a research report. This workflow finds, reads, and summarizes multiple web articles for you, presenting them in a clean table.

    image for Automate Web Research and Summarize Multiple Sources

    Workflow Overview

    What is this?

    What is this?

    This workflow acts as an automated research assistant. You provide a question or topic, and it searches the web, fetches the content from the top results, individually summarizes each article, and then presents the findings in a structured Markdown table with links to the original sources.
    What problem does it solve?

    What problem does it solve?

    It eliminates the time-consuming and tedious process of manually clicking through search results, reading long articles, and trying to synthesize key information. This tool automates the entire research and summarization cycle, freeing you up to focus on analysis and action.
    What will you get?

    What will you get?

    You get a digestible research briefing on any topic in minutes, not hours. This provides a comprehensive overview from multiple sources at a glance, allowing you to quickly grasp the main points of a subject and make faster, more informed decisions. It also creates a perfectly formatted, shareable report.

    Apps Included

    • Code
    • Jina AI
    • OpenAI

    How to Use

    Prerequisites

    Setup Steps

    1

    Configure All Required API Keys

    Ensure your API keys for OpenAI, Tavily, and Jina AI are correctly added and enabled in the 'Settings > Tools' section of your Dify account.

    2

    Import the Workflow

    Import this application into your Dify workspace. The entire research and summarization pipeline is pre-configured.

    3

    Ask Your Research Question

    In the Start Node, enter your research topic or question into the 'Question' field.

    4

    Execute and Wait for Results

    Run the application. The workflow will take some time as it needs to perform a web search and then individually read and summarize multiple web pages.

    5

    Receive Your Research Table

    The final output will be a Markdown table containing each source URL and its corresponding AI-generated summary, ready to copy and use.

    Pro Tips

    1
    Add a final LLM node after the last 'Template' node. Feed all the individual summaries into this new node and ask it to write a single 'Executive Summary' of all sources for an even higher-level overview.
    2
    Modify the first Code node to filter the URLs from Tavily. You could add logic to exclude certain domains or only include URLs that contain specific keywords to refine your source selection.

    Information

    • Published date8/15/2025
    • Last updated8/15/2025

    Platform