Sémantické vyhľadávanie v znalostnej databáze
Jednoducho vyhľadávajte a získavajte informácie z dokumentov súkromnej znalostnej databázy pomocou sémantického vyhľadávania poháňaného AI. Tok rozširuje použív...
3 min čítania
The CSV Document Search component is a versatile tool designed to facilitate intelligent searching within CSV files as part of your AI workflow. It enables users to perform text-based searches or execute pandas query expressions on CSV documents, making it highly suitable for scenarios where structured data needs to be queried and analyzed dynamically.
What Does This Component Do?
This component allows you to:
- Search within CSV files using keyword search or more advanced pandas-style query expressions.
- Filter search results by specific columns, or search across all columns if none are specified.
- Control the number of results returned by setting a maximum limit.
- Choose data sources flexibly, by selecting a CSV file from internal documents or specifying an external file URL.
- Optimize performance and efficiency with caching options, reducing the need to repeatedly load and parse large files.
This makes it especially useful for integrating structured data queries into larger AI or data processing pipelines, where automated, repeatable access to tabular data is required.
Inputs
There is no input handle for this component.
Settings
| Parameter | Description | Default/Example Value | Required |
|---|---|---|---|
| CSV Document ID | Select a CSV file from internal storage for searching. | No | |
| CSV File URL | Provide an external URL to a CSV file if not using an internal document. | No | |
| Search Columns | Specify which columns to search (comma-separated). If left blank, all columns are searched. | No | |
| Case Sensitive | Determines if the search should distinguish between uppercase and lowercase text. | False | No |
| Max Results | Sets the maximum number of results to return for each search. | 5000 | Yes |
| Cache TTL | Defines how long the CSV content should be cached (various intervals from “No cache” to “1 year”). | 2 weeks | No |
| Verbose | Enables detailed output for debugging or development purposes. | False | No |
| Tool Name | Assign a custom name to the tool for referencing within agent workflows. | No | |
| Tool Description | Provide a description for the tool to help agents understand its purpose and usage. | No |
Outputs
- Tool: The primary output is a Tool object, which can be integrated into your workflow or used by agents to perform document searches as needed.
Typical Use Cases
- Automated Data Extraction: Retrieve relevant rows from large CSV datasets based on user-provided queries or parameters.
- Data Preprocessing: Filter and extract subsets of data as a preparatory step for further analysis or machine learning.
- Dynamic Data Lookup in AI Agents: Allow AI agents to access and search tabular data on-demand as part of a broader decision-making process.
Why Use CSV Document Search?
- Flexibility: Works with both internal and external CSV files, adapts to different data storage strategies.
- Performance: Caching options enhance speed and reduce repeated data loading.
- Customizability: Search parameters, result limits, and search scope (columns, case-sensitivity) can be tailored to specific needs.
- Seamless Integration: Designed to be a modular component in larger AI workflows, facilitating structured data access for downstream tasks.
This component is ideal for anyone who needs programmatic, repeatable search capabilities over tabular data within an AI-powered automation or analytics pipeline.