Memoid JavaScript SDK

JavaScript/TypeScript SDK for Memoid - AI Memory Layer for intelligent applications.

Installation

npm install memoid
# or
yarn add memoid
# or
pnpm add memoid

Quick Start

import { MemoryClient } from "memoid";

const client = new MemoryClient("your-api-key");

// Add memory from conversation
const memories = await client.add(
  [
    { role: "user", content: "I'm vegetarian and love Italian food" },
    { role: "assistant", content: "Great! I'll remember that." },
  ],
  { user_id: "user_123" }
);

// Search memories
const results = await client.search("What dietary restrictions?", {
  user_id: "user_123",
});

results.forEach((result) => {
  console.log(`${result.memory} (score: ${result.score})`);
});

API Reference

MemoryClient

const client = new MemoryClient(apiKey, options?);

Options:

• baseUrl: API base URL (default: https://api.memoid.dev)
• timeout: Request timeout in ms (default: 30000)

add(messages, options?)

Add memories from a conversation.

const memories = await client.add(
  [{ role: "user", content: "My name is John" }],
  { user_id: "user_123" }
);

get(memoryId)

Get a specific memory by ID.

getAll(options?)

Get all memories with optional filtering.

const memories = await client.getAll({
  user_id: "user_123",
  limit: 50,
});

search(query, options?)

Search memories using natural language.

const results = await client.search("user preferences", {
  user_id: "user_123",
  limit: 10,
});

update(memoryId, memory, metadata?)

Update a memory.

delete(memoryId)

Delete a memory.

deleteAll(options?)

Delete all memories with optional filtering.

getGraph(options?)

Get the knowledge graph.

const graph = await client.getGraph({ user_id: "user_123" });
console.log(graph.entities, graph.relations);

TypeScript

Full TypeScript support with exported types:

import type { Memory, SearchResult, Graph } from "memoid";

License

MIT