Skip to main content

Interface: MemoryConfig

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:200

Top-level configuration object for the Memory facade.

All fields are optional; sensible defaults are applied per field. A minimal {} config is valid and will use a temporary SQLite brain file with graph + self-improvement enabled.

Properties​

connectionString?​

optional connectionString: string

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:227

Postgres connection string. Required when store='postgres'.

Example​

'postgresql://postgres:wunderland@localhost:5432/agent_memory'

consolidation?​

optional consolidation: ExtendedConsolidationConfig

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:280

Consolidation schedule and thresholds.

decay?​

optional decay: boolean

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:277

Whether memory traces lose strength over time following an Ebbinghaus forgetting-curve model.

Default​

true

embed()?​

optional embed: (text) => Promise<number[]>

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:256

Optional embedding function for generating vectors at remember/recall time. When provided, enables HNSW vector search in recall() and stores embeddings alongside text in remember(). Without this, recall() falls back to FTS5-only.

Parameters​

text​

string

Returns​

Promise<number[]>

Example​

const mem = await Memory.createSqlite({
  embed: async (text) => {
    const res = await openai.embeddings.create({ model: 'text-embedding-3-small', input: text });
    return res.data[0].embedding;
  },
});

embeddings?​

optional embeddings: EmbeddingConfig

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:239

Embedding model configuration (provider name + optional model).

graph?​

optional graph: boolean

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:263

Whether to build and maintain a knowledge graph alongside the vector store. When enabled, entity co-occurrence and semantic edges are tracked.

Default​

false

ingestion?​

optional ingestion: IngestionConfig

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:283

Document ingestion settings applied to all ingest() calls by default.

path?​

optional path: string

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:221

File-system path for stores that require one (e.g. SQLite db file). Ignored by in-memory and remote stores.

Example​

'./data/agent-memory.sqlite'

qdrantApiKey?​

optional qdrantApiKey: string

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:236

Qdrant API key for cloud instances. Optional.

qdrantUrl?​

optional qdrantUrl: string

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:233

Qdrant base URL. Required when store='qdrant'.

Example​

'http://localhost:6333'

selfImprove?​

optional selfImprove: boolean

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:270

Whether the agent may autonomously refine and restructure its own memories (write new insight traces, prune contradictions, merge redundancies).

Default​

false

store?​

optional store: "memory" | "postgres" | "qdrant" | "sqlite" | "neo4j" | "hnsw"

Defined in: packages/agentos/src/cognition/memory/io/facade/types.ts:214

Persistence backend for memory traces.

The Phase 1 facade currently implements the SQLite path at runtime. Other values are reserved for future backends and will throw if selected.

  • 'sqlite' – file-based SQLite (implemented; recommended).
  • 'memory' – reserved for a future in-process backend.
  • 'qdrant' – reserved for a future vector-database backend.
  • 'neo4j' – reserved for a future graph-database backend.
  • 'hnsw' – reserved for a future ANN-only backend.

Default​

'sqlite'