Skip to main content

​
Installation & Setup

MEMANTO requires a Moorcheh API key to function. This guide walks you through setup in minutes.

​
Prerequisites

  • Python 3.10+ (for CLI and local server)
  • Docker (optional, for containerized deployment)
  • Moorcheh API key (free tier available)

​
Step 1: Get a Moorcheh API Key

MEMANTO relies on Moorcheh.ai for semantic search. You need an API key to get started.

​
Create Your API Key

  1. Go to https://console.moorcheh.ai/api-keys
  2. Sign up or log in (free tier available)
  3. Click “Create New API Key”
  4. Copy the key
  5. Save it somewhere safe—never commit to git

​
Free Tier Benefits

  • 500 monthly credits = ~100,000 operations
  • No credit card required
  • No indexing delays (instant semantic search)
  • Enough for heavy development (multiple agents continuously)

​
Step 2: Install MEMANTO

Install the MEMANTO CLI to interact with memory operations: 
# Using pip
pip install memanto

# Or using uv (modern Python package manager)
pip install uv
uv tool install memanto
Verify installation: 
memanto --version

​
Option B: From Source

For development or customization: 
git clone https://github.com/moorcheh-ai/memanto.git
cd memanto

# Install dependencies
pip install uv
uv sync

# Or with pip
pip install -e .

​
Option C: Docker Installation

Build a containerized MEMANTO server: 
git clone https://github.com/moorcheh-ai/memanto.git
cd memanto

# Build image
docker build -t memanto .

# Run container
docker run -p 8000:8000 \
  -e MOORCHEH_API_KEY=mk_your_api_key \
  memanto

​
Step 3: Configure Your API Key

memanto
This prompts you to:
  1. Enter your Moorcheh API key
  2. Configure optional settings (schedule time for daily summaries, etc.)
Configuration is stored in ~/.memanto/config.json (secure, not in git).

​
Manual Configuration

If you prefer to skip the interactive setup: 
# Set environment variable
export MOORCHEH_API_KEY=mk_your_api_key

# Or create .env file in project directory
echo "MOORCHEH_API_KEY=mk_your_api_key" > .env

​
Step 4: Verify Installation

​
Test CLI Setup

memanto status
This shows:
  • API key validity ✓
  • Moorcheh connectivity ✓
  • Configuration ✓

​
Test API Server

Start the local API server: 
memanto serve
Expected output: 
INFO:     Uvicorn running on http://0.0.0.0:8000
INFO:     Application startup complete
Verify it’s running: 
curl http://localhost:8000/health
Expected response: 
{
  "status": "healthy",
  "service": "MEMANTO",
  "moorcheh_connected": true
}

​
Access Web Dashboard

Once the server is running, open the web UI: 
memanto ui
This opens an interactive dashboard at http://localhost:3000 (or similar) where you can:
  • View all agents
  • Create/activate agents
  • Store and retrieve memories
  • Monitor sessions

​
Step 5: Create Your First Agent

# Create an agent
memanto agent create my-first-agent

# Activate it (starts a session)
memanto agent activate my-first-agent

# Verify activation
memanto status
You should see: 
Active Agent:      my-first-agent
Active Session:    <session-token>
Session Expires:   <expiration-time>

​
Troubleshooting

​
”API key not found” error

# Make sure you configured the key
memanto config show

# If not set, run setup again
memanto

​
”Moorcheh connection failed” error

# Verify your API key is configured
memanto config show

# Test connectivity to Moorcheh
curl -X GET "https://api.moorcheh.ai/v1/health" \
  -H "Authorization: Bearer mk_your_api_key"

​
”Port 8000 already in use” error

Use a different port: 
memanto serve --port 8001

​
“Module not found” error

Reinstall MEMANTO: 
pip install --upgrade memanto

# Or from source
git clone https://github.com/moorcheh-ai/memanto.git
cd memanto
pip install -e .

​
Next Steps

All set! You’re ready to start building memory-enabled agents.