Skip to main content
Recipes

Memory

Memory enables agents to remember past conversations and maintain context across sessions.

Basic Memory Setup​

import { openai } from "@ai-sdk/openai";
import { Agent, Memory, VoltAgent } from "@voltagent/core";
import { LibSQLMemoryAdapter } from "@voltagent/libsql";
import { honoServer } from "@voltagent/server-hono";

const memory = new Memory({
  storage: new LibSQLMemoryAdapter({
    url: "file:./.voltagent/memory.db",
  }),
});

const agent = new Agent({
  name: "Memory Agent",
  instructions: "A helpful assistant that remembers conversations",
  model: openai("gpt-4o-mini"),
  memory,
});

new VoltAgent({
  agents: { agent },
  server: honoServer({ port: 3141 }),
});

Working Memory (Structured Context)​

Working memory lets agents maintain structured information across conversations.

JSON Schema Format​

import { z } from "zod";

const workingMemorySchema = z.object({
  userProfile: z.object({
    name: z.string().optional(),
    preferences: z.array(z.string()).optional(),
  }),
  context: z.object({
    currentGoal: z.string().optional(),
    importantNotes: z.array(z.string()).optional(),
  }),
});

const memory = new Memory({
  storage: new LibSQLMemoryAdapter(),
  workingMemory: {
    enabled: true,
    scope: "user", // or "conversation"
    schema: workingMemorySchema,
  },
});

Markdown Template Format​

const template = `
## User Profile
- Name: {name}
- Preferences: {preferences}

## Context
{context}
`;

const memory = new Memory({
  storage: new LibSQLMemoryAdapter(),
  workingMemory: {
    enabled: true,
    scope: "conversation",
    template,
  },
});
import { AiSdkEmbeddingAdapter, InMemoryVectorAdapter } from "@voltagent/core";

const memory = new Memory({
  storage: new LibSQLMemoryAdapter(),
  embedding: new AiSdkEmbeddingAdapter(openai.embeddingModel("text-embedding-3-small")),
  vector: new InMemoryVectorAdapter(),
});

Full Example​

See the complete examples:

Table of Contents