Getting Started

MemorySync Documentation

MemorySync is a memory API for AI applications: your agent writes what it learns about each user, and MemorySync returns the relevant parts later, ranked by meaning and scoped to that user. Start with one write and one retrieval, then move to the pattern that matches your product.

Start here

Run the 5-minute quickstart

Create an API key, store one memory, and retrieve it.

Understand the memory model

Learn what MemorySync stores, how scope works, and what retrieval returns.

Choose a build pattern

Apply memory to chatbots, agents, personalization, or RAG.

Look up an endpoint

Go directly to request and response contracts.

Install an SDK

Use the official Python or Node SDK with typed examples.

Operate in production

Plan reliability, observability, incidents, and deployment.

Connect a source

Import knowledge from a supported external source.

Manage plans and access

Review usage, billing, SSO, roles, and enterprise controls.

Review release notes

Track product and documentation changes over time.

The three ideas to learn first

Idea	What it means	Where to continue
Memory	Durable context your application writes to MemorySync and retrieves later.	What is a Memory
Scope	In MemorySync, organization, project, and end-user identifiers keep workloads separated.	Multi-tenant Applications
Retrieval	A natural-language query to MemorySync returns relevant memory for the same scope.	Retrieval Pipeline

Your first useful result

The shortest path to a working MemorySync integration is: create a key, choose a project, identify an end user, write a useful fact, and query that fact back.

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
    project_id=os.environ["MEMORYSYNC_PROJECT_ID"],
    end_user_id="user-123",
)

memory = client.add(
    "The user prefers concise answers and dark mode.",
    source="chat",
    metadata={'conversation_id':'conv-42'},
)

import os
from memorysync import MemorySyncClient

client = MemorySyncClient(
    api_key=os.environ["MEMORYSYNC_API_KEY"],
    base_url="https://api.memorysync.io",
    project_id=os.environ["MEMORYSYNC_PROJECT_ID"],
    end_user_id="user-123",
)

result = client.query("How should I format the answer?", k=5)
for memory in result.memories:
    print(memory.text)

Safe defaults

Keep API keys on your server and load them from environment variables.
Use a stable, non-sensitive external identifier for X-End-User-ID.
Store facts that improve future behavior—not complete prompts, secrets, or raw credentials.
Use separate projects or keys when workloads have different access boundaries.

Was this page helpful?

Why MemorySync