Getting Started with Adobe PDF Library 21 Using AI Coding Agents

Adobe PDF Library (APDFL) 21 is Datalogics's most capable PDF SDK, built on the same Adobe source code that powers Acrobat. Whether you're working in Modern C++, Adobe C/C++, .NET, .NET Framework, or Java, you can now accelerate your integration using AI coding agents — Claude Code, Codex, Cursor, and others — connected to Datalogics's Scout MCP server.

This guide walks through getting a trial license, connecting your AI agent to Scout, installing APDFL 21 for your language of choice, and putting your agent to work generating and debugging code against the SDK.

What Is the Scout MCP Server?

Datalogics exposes Scout, an AI assistant with deep knowledge of the Adobe PDF Library SDK and its documentation, as an MCP (Model Context Protocol) server. Developers using MCP-compatible clients like Claude Desktop, Claude Code, Codex, or Cursor can connect directly to Scout and query the SDK's APIs, methods, and implementation patterns — without leaving their editor.

MCP endpoint: https://datalogics.mcp.kapa.ai

Add this endpoint in your AI client's MCP server settings and authenticate as required by your client.

Once connected, you can ask Scout questions like:

• "How do I merge two PDFs in APDFL .NET?"
• "Show me how to add a digital signature in Java."
• "What's the Modern C++ way to extract text from a PDF?"

Step 1: Get Your Free Trial License Key

Before writing any code, obtain a free trial activation key:

1. Visit the Adobe PDF Library free trial page.
2. Fill out the form and submit.
3. Save your activation key — you'll need it for every language interface.

Step 2: Connect Your AI Agent to Scout

In your MCP-compatible client (Claude Code, Codex, Cursor, etc.), add the Datalogics MCP server endpoint:

https://datalogics.mcp.kapa.ai

Once connected, your AI agent can answer implementation questions, generate boilerplate, and guide you through the SDK — for any of the language interfaces below.

Step 3: Install APDFL 21 for Your Language

.NET (C# / VB.NET)

Install via NuGet:

1. In Visual Studio 2022 or VS Code, go to Tools → NuGet Package Manager → Manage NuGet Packages for Solution.
2. Search for Datalogics and select Adobe.PDF.Library.LM.NET.
3. Install and accept the license terms.

Activate your license in code:

Library.LicenseKey = "xxxx-xxxx-xxxx-xxxx";
using (Library lib = new Library())
{
    // Your APDFL code here
}

Or run a sample via the dotnet CLI:

cd ./Images/RasterizePage
dotnet build
dotnet run

Resources:

• Getting Started (.NET)
• API Reference
• C# .NET Samples on GitHub
• VB.NET Samples on GitHub

.NET Framework (C# / VB.NET)

The setup mirrors .NET above — install Adobe.PDF.Library.LM.NET via NuGet in Visual Studio 2022.

Resources:

• Getting Started (.NET Framework)
• API Reference
• .NET Framework Samples on GitHub
• VB.NET Samples on GitHub

Java (Maven)

Install via Maven:

1. Go to Datalogics Adobe PDF Library on Maven Central.
2. Copy the dependency snippet and paste it into the dependencies section of your pom.xml:

<dependency>
    <groupId>com.datalogics.pdfl</groupId>
    <artifactId>pdfl</artifactId>
    <version><!-- latest version --></version>
</dependency>

3. Provide your activation key when prompted on first run.

Resources:

• Getting Started (Java)
• API Reference
• Java Maven Samples on GitHub
• Kotlin Samples on GitHub

Modern C++

APDFL 21 introduces a NEW Modern C++ interface built on C++17, featuring:

• RAII-based resource management (no manual tracking or releasing)
• Exception-driven error handling
• std::unique_ptr / std::shared_ptr ownership models
• Strongly typed flags and a unified datalogics_interface namespace

Resources:

• Getting Started (Modern C++)
• API Reference
• Modern C++ Samples on GitHub

Adobe C/C++ (Classic Interface)

The original Adobe C/C++ interface remains fully supported in APDFL 21. Existing integrations continue unchanged — migration to Modern C++ is optional.

Resources:

• Getting Started (Adobe C/C++)
• API Reference
• C++ Samples on GitHub

Step 4: Ask Your AI Agent to Help You Build

With Scout connected via MCP, your AI coding agent can:

• Generate starter code for your chosen language.
• Explain specific API methods and classes from the API Reference.
• Help debug integration issues.
• Suggest the right APDFL approach for your PDF workflow.

Additional Resources

Next Steps

• Grab your free trial license key and pick the language interface that fits your project.
• Connect your AI client to https://datalogics.mcp.kapa.ai and start querying Scout directly from your editor.
• Contact Datalogics if you need help scoping your integration.