ManagedAgent provides agent invocation with automatic tracking and automatic judge evaluation.

The class is stateless: each run() call sends the prompt directly to the underlying Runner and returns a ManagedResult. Conversation history, if any, must be managed by the caller (or by the Runner implementation).

Obtain an instance via LDAIClient.createAgent().

Hierarchy

  • ManagedAgent

Constructors

Properties

Methods

Constructors

Properties

_logger?: LDLogger
aiAgentConfig: LDAIAgentConfig
runner: Runner

Methods

  • Invoke the agent with a prompt string and return a ManagedResult.

    run() resolves before ManagedResult.evaluations resolves. Awaiting evaluations guarantees both judge evaluation and tracker.trackJudgeResult() are complete.

    Parameters

    • prompt: string

      The user input to send to the agent.

    Returns Promise<ManagedResult>

    Promise resolving to ManagedResult (before evaluations settle).

Generated using TypeDoc