Skip to main content
Category: Utility Type: Information Retrieval

​
Overview

The Web Search Node enables your workflow to search the internet in real time. It retrieves up-to-date information, articles, summaries, and metadata from web sources based on a user-defined query. This node is ideal for workflows that need live data — such as researching topics, finding references, or generating summaries from the latest information available online.

​
Description

The Web Search Node connects your workflow to a web search engine using AI-assisted processing. When given a search query, it automatically:
  • Searches the web for the most relevant results.
  • Extracts titles, URLs, publication dates, and short descriptions (snippets).
  • Optionally provides a concise AI-generated summary of the top results.
This node helps automate research workflows, news monitoring, or any process that benefits from current web data — without leaving your automation environment.

​
Input Parameters

The Web Search Node accepts the following input parameter:
  • searchQuery (string, required) The main search term or question you want to query on the web. Example: "latest breakthroughs in quantum computing" or "top AI startups 2025". This input determines what kind of content the node retrieves and analyzes.

​
Output Parameters

After execution, the node returns structured web search data and metadata.
  • summary (string) A short AI-generated summary of the most relevant search results. Useful for quickly understanding the overall context of the search.
  • results[] (array) A list of individual search results retrieved from the web. Each result includes the following fields:
    • results[].title (string) — Title of the web page or article.
    • results[].url (string) — Direct link to the web page.
    • results[].snippet (string) — A short excerpt or description from the page content.
    • results[].date (string) — The original publication date of the content (if available).
    • results[].last_updated (string) — The last modified or indexed date of the page.
  • searchId (string) A unique identifier assigned to each web search operation. Useful for logging or debugging multiple search runs.

​
Output Type

  • Output Type: Text (Structured JSON Result)
The node returns all data in structured JSON format, allowing downstream nodes (e.g., summarization or report-generation nodes) to easily access and process web content. Example variable access syntax: 
{{webSearch.output.summary}}
{{webSearch.output.results[0].title}}
{{webSearch.output.results[0].url}}
{{webSearch.output.searchId}}

​
Example Usage

Input: 
{
  "searchQuery": "Top 10 programming languages 2025"
}
Expected Output: 
{
  "summary": "In 2025, Python, JavaScript, and Rust continue to dominate the developer community...",
  "results": [
    {
      "title": "Top Programming Languages in 2025",
      "url": "https://example.com/top-languages-2025",
      "snippet": "Python remains the most popular choice due to its versatility and AI ecosystem...",
      "date": "2025-09-10",
      "last_updated": "2025-10-20"
    }
  ],
  "searchId": "search-483912"
}

​
Example 2: Research Query

Input: 
{
  "searchQuery": "impact of climate change on agriculture"
}
Expected Output: 
{
  "summary": "Recent studies show a significant effect of rising temperatures and droughts on crop yields...",
  "results": [
    {
      "title": "Climate Change and Global Agriculture",
      "url": "https://example.org/climate-agriculture",
      "snippet": "Research indicates that food security challenges are expected to rise...",
      "date": "2025-08-15",
      "last_updated": "2025-10-22"
    }
  ],
  "searchId": "search-948172"
}

​
How to Use in a No-Code Workflow

  1. Add the Web Search Node
    • From the Utility category, drag and drop the Web Search node into your workflow canvas.
  2. Set the Input
    • In the input field, enter your search query or connect it dynamically from another node, e.g., {{userInput.output.query}}.
  3. Run the Workflow
    • The node will perform a real-time web search and retrieve the latest information.
  4. Use Outputs in Other Nodes
    • Connect the output to other nodes such as a Text Generation node for summarization or a Database node for logging search results.
  5. Access Dynamic Data
    • Use syntax like {{webSearch.output.summary}} or {{webSearch.output.results[0].url}} in downstream steps.

​
Best Practices

  • Always provide a clear and specific search query to get the most accurate results.
  • Avoid vague or overly long queries — concise phrasing yields better relevance.
  • Combine this node with AI Text Generation to summarize or interpret web findings automatically.
  • Use Logger or Database Nodes to store search results for later analysis.
  • Be mindful of the search frequency to prevent exceeding API limits or redundant requests.

​
Common Errors

  • “Missing searchQuery” Cause: The search query input was empty. Solution: Ensure you provide a valid text input under searchQuery.
  • “No results found” Cause: The query returned no relevant web pages. Solution: Rephrase or broaden your search query.
  • “Network request failed” Cause: The node could not connect to the search service. Solution: Check your internet connection or retry later.
  • “Invalid response structure” Cause: The search service returned an unexpected response. Solution: Contact support or retry with a simpler query.