Skip to content

Installation

Get Attune AI installed and configured in about 2 minutes.

Step 1: Install the Package

Choose your installation option:

pip install attune-ai[developer]

Includes: CLI tools, VSCode extension support, all workflows, local telemetry.

pip install attune-ai

Core functionality only. Add extras later as needed.

pip install attune-ai[full]

Everything: LLM providers, healthcare support, webhooks, caching.

pip install attune-ai[healthcare]

Includes: FHIR client, HL7 parsing, HIPAA audit logging.

Verify Installation

python -c "import attune; print(f'Attune AI v{attune.__version__}')"

Step 2: Configure an LLM Provider

Attune AI uses Anthropic Claude as its LLM provider.

Option A: Environment Variable (Quick)

export ANTHROPIC_API_KEY="sk-ant-..."

Option B: .env File (Persistent)

Create a .env file in your project root:

# .env
ANTHROPIC_API_KEY=sk-ant-...

The framework auto-detects .env files.

Option C: Interactive Setup

python -m attune.models.cli provider --interactive

Verify Provider

python -m attune.models.cli provider

Expected output: 

Current provider: anthropic
Available models: claude-opus-4-6, claude-sonnet-4-6, claude-haiku-4-5
API key configured

Step 3: Optional - Redis for Memory

Redis enables multi-agent coordination and session persistence. Skip this for now if you just want to try the framework.

brew install redis
brew services start redis
redis-cli ping  # Should return: PONG

docker run -d -p 6379:6379 --name attune-redis redis:alpine

The framework works without Redis (falls back to in-memory storage).

You can add Redis later when you need:

  • Multi-agent coordination
  • Session persistence
  • Pattern staging

See Redis Setup for production configuration.

Troubleshooting

"No API key configured"

# Check your environment
env | grep API_KEY

# Set it
export ANTHROPIC_API_KEY="sk-ant-..."

# Verify
python -m attune.models.cli provider --check

"ModuleNotFoundError: attune"

# Reinstall
pip install --upgrade attune-ai[developer]

# Or for development
pip install -e .[dev]

Python version too old

python --version  # Need 3.10+

# Use pyenv to manage versions
pyenv install 3.11
pyenv local 3.11

See Also

Next Step

You're installed! Continue to First Steps to run your first workflow.