atulya-embed

Atulya embedded CLI — a living algorithm for machine intelligence (MI), with local memory operations and automatic daemon management.

This package provides a simple CLI for storing and recalling memories using Atulya's memory engine. It automatically manages a background daemon for fast operations - no manual server setup required.

How It Works

atulya-embed uses a background daemon architecture for optimal performance:

1. First command: Automatically starts a local daemon (first run downloads dependencies and loads ML models - can take 1-3 minutes)
2. Subsequent commands: Near-instant responses (~1-2s) since daemon is already running
3. Auto-shutdown: Daemon automatically exits after 5 minutes of inactivity

The daemon runs on localhost:8888 and uses an embedded PostgreSQL database (pg0) - everything stays local on your machine.

Installation

pip install atulya-embed
# or with uvx (no install needed)
uvx atulya-embed --help

Quick Start

# Interactive setup (configures default profile)
atulya-embed configure

# Or set your LLM API key manually
export OPENAI_API_KEY=sk-...

# Store a memory (bank_id = "default")
atulya-embed memory retain default "User prefers dark mode"

# Recall memories
atulya-embed memory recall default "What are user preferences?"

All commands use the "default" profile unless you specify a different one with --profile or ATULYA_EMBED_PROFILE.

Commands

configure

Configure the default profile or create/update named profiles:

# Interactive setup for default profile
atulya-embed configure

# Create/update named profile with single command
atulya-embed configure --profile my-app \
  --env ATULYA_EMBED_LLM_PROVIDER=openai \
  --env ATULYA_EMBED_LLM_API_KEY=sk-xxx

# Create/update named profile interactively
atulya-embed configure --profile staging

This will:

• Let you choose an LLM provider (OpenAI, Groq, Google, Ollama)
• Configure your API key
• Set the model and memory bank ID
• Start the daemon with your configuration

memory retain

Store a memory:

atulya-embed memory retain default "User prefers dark mode"
atulya-embed memory retain default "Meeting on Monday" --context work
atulya-embed memory retain myproject "API uses JWT authentication"

memory recall

Search memories:

atulya-embed memory recall default "user preferences"
atulya-embed memory recall default "upcoming events"

Use -o json for JSON output:

atulya-embed memory recall default "user preferences" -o json

memory reflect

Get contextual answers that synthesize multiple memories:

atulya-embed memory reflect default "How should I set up the dev environment?"

bank list

List all memory banks:

atulya-embed bank list

profile

Manage configuration profiles:

# List all profiles with status
atulya-embed profile list

# Show current active profile
atulya-embed profile show

# Set active profile (persists across commands)
atulya-embed profile set-active my-app

# Clear active profile (revert to default)
atulya-embed profile set-active --none

# Delete a profile
atulya-embed profile delete my-app

daemon

Manage the background daemon:

atulya-embed daemon status    # Check if daemon is running
atulya-embed daemon start     # Start the daemon
atulya-embed daemon stop      # Stop the daemon
atulya-embed daemon logs      # View last 50 lines of logs
atulya-embed daemon logs -f   # Follow logs in real-time
atulya-embed daemon logs -n 100  # View last 100 lines

Configuration

Interactive Setup

Run atulya-embed configure for a guided setup that saves to ~/.atulya/embed.

Environment Variables

Variable	Description	Default
ATULYA_EMBED_PROFILE	Profile name to use (overrides active profile)	None (uses default profile)
ATULYA_EMBED_LLM_API_KEY	LLM API key (or use OPENAI_API_KEY)	Required
ATULYA_EMBED_LLM_PROVIDER	LLM provider (openai, groq, google, ollama)	openai
ATULYA_EMBED_LLM_MODEL	LLM model	gpt-4o-mini
ATULYA_EMBED_BANK_ID	Default memory bank ID (optional, used when not specified in CLI)	default
ATULYA_EMBED_API_URL	Use external API server instead of starting local daemon	None (starts local daemon)
ATULYA_EMBED_API_TOKEN	Authentication token for external API (sent as Bearer token)	None
ATULYA_EMBED_API_DATABASE_URL	Database URL for daemon	pg0://atulya-embed
ATULYA_EMBED_DAEMON_IDLE_TIMEOUT	Seconds before daemon auto-exits when idle	300

Using an External API Server:

To connect to an existing Atulya API server instead of starting the local daemon:

export ATULYA_EMBED_API_URL=http://your-server:8000
export ATULYA_EMBED_API_TOKEN=your-api-token  # Optional, if API requires auth
atulya-embed memory recall default "query"

Custom Database:

To use an external PostgreSQL database instead of the embedded pg0 database (useful when running as root or in containerized environments):

export ATULYA_EMBED_API_DATABASE_URL=postgresql://user:password@localhost:5432/dbname
atulya-embed daemon start

Note: All banks share a single database. Bank isolation happens within the database via the bank_id parameter passed to CLI commands.

Configuration Profiles

Profiles let you maintain multiple independent configurations (e.g., different API endpoints, LLM providers, or projects). Each profile runs its own daemon on a unique port (8889-9888).

The Default Profile:

When you run atulya-embed configure without specifying a profile, it configures the "default" profile. This uses the backward-compatible configuration at ~/.atulya/embed and runs on port 8888.

Creating Named Profiles:

# Create a profile with single command
atulya-embed configure --profile my-app \
  --env ATULYA_EMBED_LLM_PROVIDER=openai \
  --env ATULYA_EMBED_LLM_API_KEY=sk-xxx \
  --env ATULYA_EMBED_LLM_MODEL=gpt-4o-mini

# Create a profile interactively
atulya-embed configure --profile staging

Using Profiles:

# Option 1: Environment variable (recommended for apps)
ATULYA_EMBED_PROFILE=my-app atulya-embed memory retain default "text"

# Option 2: CLI flag
atulya-embed --profile my-app memory recall default "query"

# Option 3: Set as active (persists across commands)
atulya-embed profile set-active my-app
atulya-embed memory recall default "query"  # Uses my-app profile

# Clear active profile (revert to default)
atulya-embed profile set-active --none

Profile Management:

# List all profiles with status
atulya-embed profile list

# Show active profile
atulya-embed profile show

# Delete a profile
atulya-embed profile delete my-app

Profile Resolution Priority:

1. ATULYA_EMBED_PROFILE environment variable (highest)
2. --profile CLI flag
3. Active profile from ~/.atulya/active_profile file
4. Default profile (lowest)

Note: If a profile is specified but doesn't exist, the command will fail with an error. Profiles must be explicitly created using atulya-embed configure --profile <name>.

Files

Default Profile:

Path	Description
~/.atulya/embed	Configuration file for default profile
~/.atulya/daemon.log	Daemon logs for default profile
~/.atulya/daemon.lock	Daemon lock file (PID) for default profile

Named Profiles:

Path	Description
~/.atulya/profiles/<name>.env	Configuration file for profile
~/.atulya/profiles/<name>.log	Daemon logs for profile
~/.atulya/profiles/<name>.lock	Daemon lock file (PID) for profile
~/.atulya/profiles/metadata.json	Profile metadata (ports, timestamps)
~/.atulya/active_profile	Active profile name (when set with profile set-active)

Use with AI Coding Assistants

This CLI is designed to work with AI coding assistants like Claude Code, Cursor, and Windsurf. Install the Atulya skill:

npx skills add https://github.com/eight-atulya/atulya --skill atulya-docs

This will configure the LLM provider and install the skill to your assistant's skills directory.

Troubleshooting

Daemon won't start:

# Check logs for errors
atulya-embed daemon logs

# Stop any stuck daemon and restart
atulya-embed daemon stop
atulya-embed daemon start

Slow first command: This is expected - the first command needs to download dependencies, start the daemon, and load ML models. First run can take 1-3 minutes depending on network speed. Subsequent commands will be fast (~1-2s).

Change configuration:

# Re-run configure (automatically restarts daemon)
atulya-embed configure

License

Apache 2.0