Skip to content

Quick Start

AutoMem runs as a separate service. Choose where it runs, then connect your agent.

01 · WHERE SHOULD AUTOMEM RUN?

Bring Docker Compose (plus Git for the guided installer).

You keep the machine running and back up its volumes.

Use the prebuilt Docker image

GUIDED INSTALL · NODE + NPM

npx @verygoodplugins/mcp-automem install --target local

The guided local installer builds from source. The prebuilt image path above skips that build.

Bring An InstaPods or Railway account.

The provider runs the infrastructure. You manage the deployment and credentials.

InstaPods or Railway

See guided cloud setup

RUN THE INSTALLER · NODE + NPM

npx @verygoodplugins/mcp-automem install --target cloud

The installer prompts for a hosting provider.

Bring The AutoMem API URL and its token, if required.

The current deployment keeps its storage and embedding settings.

Check your connection

RUN THE INSTALLER · NODE + NPM

npx @verygoodplugins/mcp-automem install --target existing

Enter the URL and token when prompted.

Run the command for your selected route above. You need Node.js 20.19+, 22.13+, or 24+ and npm. The installer asks for your endpoint, lets you select agents, then shows a plan before applying it.

For a prebuilt Docker setup, enter http://localhost:8001 and your AUTOMEM_API_TOKEN using Existing Endpoint. Hosted users can choose InstaPods or Railway.

The optional welcome conversation can demonstrate the first memory tool calls. Fully quit and reopen your agent after installation. Need help with a prompt? See the installer walkthrough.

In your connected agent, send:

Chat 1 · Save a memory
Use AutoMem to remember that my demo workspace is called Lantern.

Look for a successful store_memory call. Then open a fresh chat and send:

Chat 2 · Recall it
Use AutoMem to look up the name of my demo workspace.

✓ First memory recalled. You’re connected when recall_memory returns Lantern in the new chat.

If the result is empty, give background indexing a moment and retry. Check that both chats use the same endpoint. For connection errors, see health and token checks or your agent’s platform guide.

Add more agents whenever you like. Connect them to the same AutoMem endpoint to share memories.