> ## Documentation Index
> Fetch the complete documentation index at: https://docs.keyflow.space/llms.txt
> Use this file to discover all available pages before exploring further.

# Sentence Search

### **Overview:**

The `Sentence Search` block is an efficient tool designed to process input text line-by-line and identify the lines that are most relevant to a specified search query. This feature allows users to quickly retrieve important information from large text bodies without the need for manual reading.

***

### Inputs & Outputs

| I/O | Feature | Type | Simple Explanation |
| - | - | - | - |
| input | `text` | string | The larger body of text where the block will perform its search for the given query. |
| input | `query` | string | The phrase or keyword you wish to find within the provided text. |
| input | `num_results` | number | Specifies the maximum number of search results to return; defaults to 5 if not defined. |
| output | `search_result` | string | Returns lines from the input text that closely match your specified query. |

***

### Use Cases

Here are some scenarios where this block can be particularly useful:

* **Information Retrieval:** When conducting research, this block helps locate specific details or quotes across extensive articles or documents without needing to sift through everything manually.
* **Customer Support:** Support teams can use this tool to find relevant responses in help center articles based on customer inquiries, enhancing response efficiency.
* **Content Review:** If you're reviewing reports or documentation drafts, leverage this feature to quickly pull pertinent feedback related directly to key topics addressed.
* **Data Analysis:** In data annotation tasks, utilize this block for extracting critical insights from lengthy datasets by searching for keywords linked with specific analysis points.

In summary, whenever there’s a need for quick and effective extraction of relevant information based on defined criteria, look no further than the `Sentence Search` block!


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.