@launchdarkly/server-sdk-ai - v2.0.21
    Preparing search index...

    Interface LDAIAgentConfig

    Agent-specific AI Config with instructions.

    interface LDAIAgentConfig {
        createTracker: () => LDAIConfigTracker;
        enabled: boolean;
        evaluator: Evaluator;
        instructions?: string;
        judgeConfiguration?: LDJudgeConfiguration;
        key: string;
        model?: LDModelConfig;
        provider?: LDProviderConfig;
        tools?: { [toolName: string]: LDTool };
    }

    Hierarchy (View Summary)

    Index
    createTracker enabled evaluator instructions? judgeConfiguration? key model? provider? tools?
    createTracker: () => LDAIConfigTracker

    Creates a new tracker for a fresh AI run. Each call mints a new runId (a UUIDv4) that LaunchDarkly uses to correlate the run's events in metrics views. Call this once per AI run; metrics from different runIds cannot be combined.

    enabled: boolean

    Whether the configuration is enabled.

    evaluator: Evaluator

    Evaluator for this agent config. Populated at evaluation time. Not part of the flag value shape.

    instructions?: string

    Instructions for the agent.

    judgeConfiguration?: LDJudgeConfiguration

    Judge configuration for AI Configs being evaluated. References judge AI Configs that should evaluate this AI Config.

    key: string

    The key of the AI Config.

    model?: LDModelConfig

    Optional model configuration.

    provider?: LDProviderConfig

    Optional configuration for the provider.

    tools?: { [toolName: string]: LDTool }

    Root-level tools map keyed by tool name. Distinct from model.parameters.tools[].

    Settings
    Member Visibility

    On This Page
    Properties