Skip to main content

Interface: InfiniteContextConfig

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:16

Properties​

compactionThreshold​

compactionThreshold: number

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:22

Trigger compaction when context reaches this fraction of max tokens (0–1).

enabled​

enabled: boolean

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:18

Enable infinite context window management.

llmInvoker()?​

optional llmInvoker: (prompt) => Promise<string>

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:43

Writes the summaries. CognitiveMemoryManager falls back to the reflector's, the observer's or the feature-detection invoker; without any, and when a call fails or answers nothing, compaction writes extractive summaries.

Parameters​

prompt​

string

Returns​

Promise<string>

logRetention​

logRetention: number

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:33

Max compaction log entries retained in memory.

maxSummaryChainTokens​

maxSummaryChainTokens: number

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:35

Token budget for the rolling summary chain header.

modelId?​

optional modelId: string

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:45

Model ID for summarization calls.

preserveRecentTurns​

preserveRecentTurns: number

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:29

Turns kept as they are, newest first (two messages a turn), while older messages remain to summarize. A conversation shorter than that, over its threshold all the same, gives up its two oldest messages; the newest message is never summarized.

strategy​

strategy: CompactionStrategy

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:20

Compaction strategy to use.

targetCompressionRatio​

targetCompressionRatio: number

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:37

Target compression ratio for summaries (e.g. 8 = 8:1).

transparencyLevel​

transparencyLevel: TransparencyLevel

Defined in: packages/agentos/src/cognition/memory/pipeline/context/types.ts:31

Transparency logging level.