MemSDK

MemSDK extracts Supermemory’s public API surface into a backend-agnostic TypeScript contract. Application code can target one memory interface, then swap adapters without rewriting its memory layer.

“Oh wow that is awesome.” Dhravya Shah, founder of Supermemory

The problem

Every AI memory backend ships a different SDK. Moving from Supermemory to Letta to another backend usually means rewriting document add, search, forget, and list workflows even when the product behavior is conceptually the same.

The approach

MemSDK freezes a proven interface instead of inventing a new one. It provides:

SupermemoryInterface as a type-level contract.
Zod schemas for runtime request validation.
Adapters that prove the contract can map to real backends.

Repositories

Repository	Purpose
memsdk	Core interface and schemas
memsdk-letta	Letta-backed implementation
memsdk-worlds	Worlds-backed implementation
memsdk-e2e	Private conformance scenarios across backends

Installation

Both memsdk and its adapters are distributed directly from GitHub. They are not currently published to the npm registry. Install the core package with any npm-compatible package manager:

npm install github:wazootech/memsdk

pnpm add github:wazootech/memsdk

yarn add github:wazootech/memsdk

bun add github:wazootech/memsdk

Install a backend adapter (e.g. Letta):

npm install github:wazootech/memsdk-letta

pnpm add github:wazootech/memsdk-letta

yarn add github:wazootech/memsdk-letta

bun add github:wazootech/memsdk-letta

For reproducible installs, pin to a tag or commit:

npm install github:wazootech/memsdk#<tag-or-commit>

pnpm add github:wazootech/memsdk#<tag-or-commit>

yarn add github:wazootech/memsdk#<tag-or-commit>

bun add github:wazootech/memsdk#<tag-or-commit>

Packages build from source during installation via prepare, then expose compiled ESM entrypoints and TypeScript declarations from dist.

Runtime support

memsdk is plain ESM compiled with tsc. Zod is the only runtime dependency, and the package declares no engines constraint. Runtime support therefore comes down to how the package is resolved, not runtime-specific code:

Runtime	Status	Resolution path
Node.js	Supported	Any npm-compatible package manager (`npm`, `pnpm`, `yarn`)
Bun	Supported	`bun add github:wazootech/memsdk`
Vite/browser	Supported through bundler	Package dependency, then import normally from app code
Edge runtimes	Supported through bundler	Package dependency, bundled by the platform deployer
Deno	Not first-class yet	Blocked by distribution channel, not the runtime: `npm:` specifiers need an npm-registry artifact
Browser/CDN	Not first-class yet	Blocked by distribution channel: esm.sh/jsdelivr/unpkg resolve npm packages

Because memsdk is primarily a TypeScript contract plus Zod schemas, browser and edge use should go through a bundler today. Direct <script>/CDN usage is not a supported distribution path yet. Publishing to the npm registry would unblock Deno npm: specifiers, browser CDNs, and edge registries from a single artifact. Adapters such as memsdk-letta depend on backend SDKs and should run server-side unless browser bundling has been validated for your app.

Usage

Use the core contract to write backend-agnostic app code:

import type { SupermemoryInterface } from "memsdk";

function buildApp(client: SupermemoryInterface) {
  return client.search({ q: "project notes" });
}

search is a callable top-level method on the contract. The legacy search.documents, search.execute, and search.memories variants are deprecated; use client.search() for v4 memory search.

Raw API to typed results

Before: a raw HTTP call against a memory backend, with an untyped response shape you must match by hand:

const response = await fetch(`${baseUrl}/memories/search`, {
  method: "POST",
  headers: { "Content-Type": "application/json" },
  body: JSON.stringify({ q: "project notes" }),
});
const data = await response.json(); // untyped: shape depends on the backend

After: the same call through the MemSDK contract, with request validation and typed results:

import { LettaMemoryClient } from "memsdk-letta";

const client = new LettaMemoryClient({
  baseUrl: "http://localhost:8283",
  apiKey: "sk-your-api-key",
});

const results = await client.search({ q: "project notes" });
// results: typed SearchResults from the contract

Wire in an adapter

import { LettaMemoryClient } from "memsdk-letta";

const client = new LettaMemoryClient({
  baseUrl: "http://localhost:8283",
  apiKey: "sk-your-api-key",
});

await client.add({
  content: "Dhravya prefers ML over traditional programming.",
  containerTag: "user_123",
});

const results = await client.search({
  q: "ML",
  containerTag: "user_123",
});
console.log(results.results);
// [{ id: "mem_123", similarity: 0.93, memory: "Dhravya prefers ML over traditional programming.", ... }]

Status

MemSDK is useful when you want memory portability more than hosted-memory lock-in. The public surface focuses on memory-domain methods first: add, profile, documents, search, and memories.

Welcome

Platform

Skills

Ecosystem

Worlds

Guides

Console

Wazoo CLI

Integrations

Comparisons

Open source

Platform API

Worlds API

The problem

The approach

Repositories

Installation

Runtime support

Usage

Raw API to typed results

Wire in an adapter

Status

​The problem

​The approach

​Repositories

​Installation

​Runtime support

​Usage

​Raw API to typed results

​Wire in an adapter

​Status

The problem

The approach

Repositories

Installation

Runtime support

Usage

Raw API to typed results

Wire in an adapter

Status