Web Search Node
Perform web searches and retrieve information
What is a Web Search node?
A Web Search node allows you to perform searches on the internet and retrieve relevant information. This node can be particularly useful for gathering data from various online sources, which can then be utilized in your workflows. The results obtained from the Web Search node can be formatted and processed further as needed.
Knowledge Base Editor
If you click the settings icon in the node, you can open the Knowledge Base Editor. This allows you to manage the settings of the Web Search node directly within the workflow.
Settings
The Web Search node has several configurable settings:
- Chunking Algorithm: This setting allows you to choose how the search results are chunked.
- Chunk Length: You can specify the maximum length of each chunk of information retrieved.
- Chunk Overlap: This setting determines how much overlap there is between chunks.
- Advanced Data Extraction: This option enables advanced features for extracting data.
- Text in images (OCR): This setting allows the node to extract text from images using Optical Character Recognition (OCR).
How to use it?
To utilize the Web Search node, you must establish connections to both its input and output edges:
- Input: This node requires a text input, typically a search query. You would connect the input to a user input node or another relevant source.
- Output: The node outputs formatted search results, which can be connected to other nodes for further processing, typically LLM nodes.
Example Output
When a search is performed, the Web Search node can return results such as:
- Title: Lionel Messi Biography
- Content: A brief overview of Lionel Messi’s career and achievements.
This output can be formatted and utilized in various ways within your workflow.