AxRLMConfig Generated TypeScript API reference. typescript api api/reference build/apidocs/Interface.AxRLMConfig.md interface AxRLMConfig

AxRLMConfig

Defined in: https://github.com/ax-llm/ax/blob/5b28f9093bb70863b59459bb6df5062d005bce41/src/ax/agent/rlm.ts#L465

RLM configuration for AxAgent.

Properties

PropertyTypeDescription
actorTurnCallback?AxAgentActorTurnCallbackCalled after each Actor turn is recorded with both raw runtime output and the formatted action-log output.
agentStatusCallback?(message: string, status: "success" | "failed") => void | Promise<void>Called when the actor signals task progress via reportSuccess(message) or reportFailure(message).
contextFieldsstring[]Input fields holding long context (will be removed from the LLM prompt).
contextPolicy?AxContextPolicyConfigContext replay, checkpointing, and runtime-state policy.
maxBatchedLlmQueryConcurrency?numberMaximum parallel llmQuery calls in batched mode (default: 8).
maxRuntimeChars?numberMaximum characters to keep from runtime output and console/log replay (default: 3000).
maxSubAgentCalls?numberGlobal cap on recursive sub-agent calls across all descendants (default: 100).
maxTurns?numberMaximum Actor turns before forcing Responder (default: 10).
onContextEvent?AxAgentOnContextEventCalled when AxAgent measures context pressure or changes compacted context state. Intended for observability; callback failures are ignored.
promptLevel?"default" | "detailed"Actor prompt verbosity and scaffolding level (default: ‘default’).
runtime?AxCodeRuntimeCode runtime for the REPL loop (default: AxJSRuntime).
summarizerOptions?Omit<AxProgramForwardOptions<string>, "functions">Default options for the internal checkpoint summarizer.
Docs