Skip to main content

Interface: GMIBaseConfig

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:246

Base configuration required to initialize a GMI instance.

Interface​

GMIBaseConfig

Properties​

beforeModelCall()?​

optional beforeModelCall: (context) => void | ChatMessage[] | Promise<void | ChatMessage[]>

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:285

Called for every model attempt after its prompt is built and before it is sent, a fallback hop's rebuilt prompt included. Messages it returns replace that attempt's prompt and do not enter the history; a hook that throws or returns an empty list is recorded on the reasoning trace and the built prompt is sent. agent({ runtime: 'gmi' }) routes onBeforeGeneration here.

Parameters​

context​

GMIModelCallContext

Returns​

void | ChatMessage[] | Promise<void | ChatMessage[]>

cognitiveMemory?​

optional cognitiveMemory: ICognitiveMemoryManager

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:254

Cognitive memory system (personality-affected encoding/retrieval with Ebbinghaus decay).

completionGateway?​

optional completionGateway: CompletionGateway

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:277

One model layer for the turn. When set, the GMI resolves the hop through it before it builds the prompt, streams through it, and moves to the next hop when an attempt fails before any output with a retryable error. llmProviderManager is then a GatewayProviderManager (see src/api/runtime/gatewayProviderManager.ts).

customSettings?​

optional customSettings: Record<string, any>

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:269

defaultLlmModelId?​

optional defaultLlmModelId: string

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:256

defaultLlmProviderId?​

optional defaultLlmProviderId: string

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:255

defaultReasoningTraceMaxEntries?​

optional defaultReasoningTraceMaxEntries: number

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:266

Runtime default for the reasoning trace's entry cap when the persona sets no reasoningTraceConfig.maxEntries. Defaults to 500.

defaultReasoningTraceMaxMessageLength?​

optional defaultReasoningTraceMaxMessageLength: number

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:268

Runtime default for the characters kept per trace message. Defaults to 1000.

learning?​

optional learning: ILearnerSession

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:298

The session's learner session, passed by a host whose learning option is on. The GMI asks it once per user turn, before the tool loop, for the personality section (a system prompt at priority 2, after the persona's base prompt) and tells each tool call the turn's learner and mode (sessionData.learning). A session that fails or does not answer within 2,000 ms never fails the turn. The host that passes the session ends it.

llmProviderManager​

llmProviderManager: AIModelProviderManager

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:249

maxToolLoopIterations?​

optional maxToolLoopIterations: number

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:261

Maximum number of tool-loop iterations before the safety break engages. Prevents runaway tool loops in processTurnStream(). Defaults to 5.

personaTraits?​

optional personaTraits: object

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:306

The persona's traits as the GMI renders them itself, when it has no learner session or the session did not answer: authored is the registry persona's traits and overlay a workflow role overlay's patch. They are read once, at initialize(), never from the persona copy setPersonalityTrait() replaces. Honesty-humility and emotionality always come from authored; an overlay that names either is not applied to it, and the trace says so. Without this field the GMI renders no personality section of its own.

authored​

authored: HexacoTraitValues

overlay?​

optional overlay: Partial<Record<"honesty" | "emotionality" | "extraversion" | "agreeableness" | "conscientiousness" | "openness", number>>

promptEngine​

promptEngine: IPromptEngine

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:248

retrievalAugmentor?​

optional retrievalAugmentor: IRetrievalAugmentor

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:252

shutdownTimeoutMs?​

optional shutdownTimeoutMs: number

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:291

How long shutdown() waits, in milliseconds, for the turns it stopped to end before it closes the GMI's memories anyway. Defaults to 8000. GMIManager passes its own shutdownTimeoutMs.

toolOrchestrator​

toolOrchestrator: IToolOrchestrator

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:251

utilityAI​

utilityAI: IUtilityAI

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:250

workingMemory​

workingMemory: IWorkingMemory

Defined in: packages/agentos/src/cognition/substrate/IGMI.ts:247