PdfContentExtractionAgentTools Class

Provides AI agent tools for extracting content from PDF documents. Handles text, image, and table extraction operations.

Inheritance
System.Object
AgentToolBase
AgentToolBase<PdfDocumentBase>
PdfContentExtractionAgentTools
Inherited Members
Namespace: Syncfusion.AI.AgentTools.PDF
Assembly: Syncfusion.DocumentSDK.AI.AgentTools.dll
Syntax
public class PdfContentExtractionAgentTools : AgentToolBase<PdfDocumentBase>

Constructors

PdfContentExtractionAgentTools(DocumentStorageManager)

Initializes a new instance of the PdfContentExtractionAgentTools class (Mode 2 � DocumentStorage).

Declaration
public PdfContentExtractionAgentTools(DocumentStorageManager manager)
Parameters
Type Name Description
DocumentStorageManager manager

The document storage manager.

PdfContentExtractionAgentTools(PdfDocumentManager)

Initializes a new instance of the PdfContentExtractionAgentTools class (Mode 1 � InMemory).

Declaration
public PdfContentExtractionAgentTools(PdfDocumentManager manager)
Parameters
Type Name Description
PdfDocumentManager manager

The PDF document manager.

Methods

ExtractImages(String, Int32, Int32, String)

Extracts images from a PDF document using a page range or the entire document. Saves the extracted images to the specified output folder.

Declaration
[Tool(Name = "ExtractImages", Description = "Extracts images from a PDF document using a page range or the entire document. documentIdOrFilePath: The document ID (InMemory mode) or input file path (DocumentStorage mode).")]
public AgentToolResult ExtractImages([ToolParameter(Description = "The document ID (InMemory mode) or input file path (DocumentStorage mode)")] string documentIdOrFilePath, [ToolParameter(Description = "Starting page index (0-based). Use -1 to extract all pages.")] int startPageIndex = -1, [ToolParameter(Description = "Ending page index (0-based). Use -1 to extract all pages.")] int endPageIndex = -1, [ToolParameter(Description = "Optional output folder path.")] string outputPath = null)
Parameters
Type Name Description
System.String documentIdOrFilePath

The document ID (InMemory mode) or input file path (DocumentStorage mode).
System.Int32 startPageIndex

Starting page index (0-based). Use -1 to extract all pages.
System.Int32 endPageIndex

Ending page index (0-based). Use -1 to extract all pages.
System.String outputPath

Optional output folder path.
Returns
Type Description
AgentToolResult

Result containing the image count and output folder path.
Remarks

Supports:

Page indexes are zero-based.

ExtractText(String, Int32, Int32)

Extracts text content from a PDF document, either from a specific range of pages or from the entire file.

Declaration
[Tool(Name = "ExtractText", Description = "Extracts text content from a PDF document, either from a specific range of pages or from the entire file. documentIdOrFilePath: The document ID (InMemory mode) or input file path (DocumentStorage mode).")]
public AgentToolResult ExtractText([ToolParameter(Description = "The document ID (InMemory mode) or input file path (DocumentStorage mode)")] string documentIdOrFilePath, [ToolParameter(Description = "Starting page index (0-based), -1 for all pages")] int startPageIndex = -1, [ToolParameter(Description = "Ending page index (0-based), -1 for all pages")] int endPageIndex = -1)
Parameters
Type Name Description
System.String documentIdOrFilePath

The document ID (InMemory mode) or input file path (DocumentStorage mode).
System.Int32 startPageIndex

Starting page index (0-based), -1 for all pages.
System.Int32 endPageIndex

Ending page index (0-based), -1 for all pages.
Returns
Type Description
AgentToolResult

Result containing the extracted text and page count.

FindTextInPdf(String, String[])

Declaration
[Tool(Name = "FindTextInPdf", Description = "Searches a PDF document for matching array of text and returns all occurrences grouped by page. documentIdOrFilePath: The document ID (InMemory mode) or input file path (DocumentStorage mode).")]
public AgentToolResult FindTextInPdf([ToolParameter(Description = "The document ID (InMemory mode) or input file path (DocumentStorage mode)")] string documentIdOrFilePath, [ToolParameter(Description = "The array of text to be searched")] string[] texts)
Parameters
Type Name Description
System.String documentIdOrFilePath
System.String[] texts
Returns
Type
AgentToolResult

GetPdfDocumentPageCount(String)

Retrieves the number of pages in a PDF document identified by the specified document ID or file path.

Declaration
[Tool(Name = "GetPdfDocumentPageCount", Description = "Returns the number of pages in the specified PDF document. documentIdOrFilePath: The document ID (InMemory mode) or input file path (DocumentStorage mode).")]
public AgentToolResult GetPdfDocumentPageCount([ToolParameter(Description = "The document ID (InMemory mode) or input file path (DocumentStorage mode)")] string documentIdOrFilePath)
Parameters
Type Name Description
System.String documentIdOrFilePath

The document ID (InMemory mode) or input file path (DocumentStorage mode).
Returns
Type Description
AgentToolResult

An AgentToolResult containing the page count of the specified PDF document if successful; otherwise, a failure result with an error message.

GetPdfDocumentPageSize(String, Int32)

Declaration
[Tool(Name = "GetPdfDocumentPageSize", Description = "Gets the width and height of a specific page in a PDF document. Works for both newly created and loaded documents without reloading.This is required when placing elements like signatures, stamps, or annotations at specific positions such as bottom-right, bottom-left, top-right, top-left.")]
public AgentToolResult GetPdfDocumentPageSize([ToolParameter(Description = "The document ID (InMemory mode) or input file path.")] string documentIdOrFilePath, [ToolParameter(Description = "Page number (1-based index).")] int pageNumber)
Parameters
Type Name Description
System.String documentIdOrFilePath
System.Int32 pageNumber
Returns
Type
AgentToolResult