Skip to content

Developer Guide - Agentic AI

This guide provides step-by-step instructions to set up and run the Agentic AI platform. Follow the steps below to clone the repositories, configure the server, and set up Agent Studio.

1. Clone the Repositories

To begin, clone the following three repositories from GitHub:

  1. XiansAI.Server
    Repository URL: https://github.com/XiansAiPlatform/XiansAi.Server.git
    Command to clone:

    Bash
    git clone https://github.com/XiansAiPlatform/XiansAi.Server.git

  2. XiansAI.Lib
    Repository URL: https://github.com/XiansAiPlatform/XiansAi.Lib.git
    Command to clone:

    Bash
    git clone https://github.com/XiansAiPlatform/XiansAi.Lib.git

  3. Agent Studio
    Repository URL: https://github.com/XiansAiPlatform/agent-studio.git
    Command to clone:

    Bash
    git clone https://github.com/XiansAiPlatform/agent-studio.git

2. Setting Up the Server

To configure and run the server, follow these steps:

  1. Obtain Configuration Files

    Request the appsettings.json and appsettings.Development.json files from a team member. These files contain the necessary configuration settings for the server.

  2. Place Configuration Files

    Copy the appsettings.json and appsettings.Development.json files into the src folder of the XiansAI.Server repository.

    Example path:

    Text Only
    XiansAI.Server/src/

  3. Run the Server

    Follow the specific instructions provided in the XiansAI.Server repository's README file to build and run the server.

3. Setting Up Agent Studio

To configure and run Agent Studio, follow these steps (see also the full Agent Studio installation guide):

  1. Create an environment file

    • Navigate to the root directory of the agent-studio repository.
    • Copy the example env file:
      Bash
      cp .env.example .env.local

  2. Configure environment variables

    • Edit .env.local and set at least:
      • XIANS_SERVER_URL — your local server URL (e.g. http://localhost:5005)
      • XIANS_APIKEY — an API key from the server (bootstrap or tenant settings)
      • NEXTAUTH_URL — http://localhost:3010
      • NEXTAUTH_SECRET — generate with openssl rand -base64 32
      • At least one OAuth/OIDC provider (Google, Azure AD, Keycloak, or Visma Connect)

  3. Install Dependencies

    • Run the following command to install all required dependencies:
      Bash
      npm install

  4. Start Agent Studio

    • Start the development server by running:

      Bash
      npm run dev

    • Agent Studio should now be accessible at http://localhost:3010.

4. Setting Up the Library (XiansAI.Lib)

To configure and use the XiansAI.Lib library, follow these steps:

  1. Run the Library

  2. After cloning the XiansAI.Lib repository, navigate to its root directory and run the following command to build and execute the library in release mode:

    Bash
    dotnet run -c Release

  3. Making Changes to the Library

  4. If you make any changes to the XiansAI.Lib codebase, you need to update its reference in the agent's package dependency list:
  5. Locate the .csproj file of the XiansAI.Lib project.
  6. Copy the file path of the .csproj file.

  7. Update the agent's package dependency list to reference the updated .csproj file location.

    Example of referencing the .csproj file in another project:

    XML
    <ProjectReference Include="path/to/XiansAI.Lib.csproj" />

  8. Rebuild the Dependent Project

  9. After updating the reference, rebuild the dependent project to ensure the changes in the library are applied.

Notes

  • Ensure you have the required versions of .NET SDK installed on your system before setting up the library. Refer to the XiansAI.Lib repository's README for version requirements.
  • For the server and library, ensure you have the necessary runtime environment (e.g., .NET Core) installed. Check the respective repositories for specific prerequisites.
  • Agent Studio requires Node.js 20+ (>=20.9.0).
  • If you encounter any issues during setup, consult the respective repository's documentation or reach out to your team for assistance.

By following these steps, you should be able to successfully set up and run the Agentic AI platform.