Gemini CLI Setup

Configure MemPad with Google's Gemini CLI for persistent AI memory.

Prerequisites

Gemini CLI installed
A MemPad account (request access if you don't have one)

Installation

Add MemPad to your Gemini CLI configuration:

# Create or edit ~/.gemini/settings.json
{
  "mcpServers": {
    "mempad": {
      "url": "https://api.mempad.ai/mcp"
    }
  }
}

Gemini CLI has native Streamable HTTP and OAuth 2.1 support - no extra configuration needed.

Authentication

The first time you use MemPad, Gemini CLI will open your browser:

A browser window opens automatically
Sign in with your MemPad email and password
Approve the authorization request
Return to your terminal - you're authenticated!

Usage

Once configured, use these commands with Gemini:

mp!

Load your memory into the conversation

mp:

Save the current context to your memory

Example Session

You: mp!
Gemini: I've loaded your memory. You prefer Python with type hints,
you're building a FastAPI backend, and you like detailed explanations.

You: Add that I'm now using Pydantic v2
Gemini: Updated! I've added Pydantic v2 to your preferences. mp:

Memory saved.

Troubleshooting

Config File Location

The settings file should be at ~/.gemini/settings.json. Create the directory if it doesn't exist:

mkdir -p ~/.gemini

Authentication Issues

If authentication fails, check that the URL is exactly https://api.mempad.ai/mcp (no trailing slash).