Quickstart: attune CLI

Run your first attune command and route user input to a skill in under a minute.

attune version

Expected output:

attune x.y.z

Prerequisites

• attune is installed and available on your PATH
• You have a terminal open in your project directory

Step 1: Verify your setup

Run the doctor command to confirm everything is configured correctly:

attune doctor

If the output shows no errors, you're ready to proceed.

Step 2: Check today's costs

attune costs today

This calls cmd_costs_today and prints a summary of your AI usage costs for the current day.

Step 3: Route user input programmatically

Use route_user_input to send a plain-language string to the appropriate skill:

from attune.cli_router import route_user_input

result = route_user_input("show me my lessons")
print(result)

Expected output is a dict describing the routed skill and any arguments resolved from your input.

Step 4: Teach the router a custom shortcut

from attune.cli_router import HybridRouter

router = HybridRouter()
router.learn_preference(keyword="costs", skill="cmd_costs_today")
result = router.route("costs")
print(result)

learn_preference stores a RoutingPreference entry (with fields keyword, skill, args, usage_count, and confidence) so the router recognises your shorthand on every future call.

What you just did

• Confirmed your installation with cmd_doctor
• Viewed a cost summary with cmd_costs_today
• Routed plain-language input with route_user_input
• Registered a custom routing shortcut with HybridRouter.learn_preference

Next: say "how do I configure routing preferences?" for a full walkthrough of HybridRouter, RoutingPreference fields, and persistent preference storage.