Getting started

Step 2

Connect your WHOOP

The setup command opens a browser window for WHOOP OAuth authentication. Your credentials are stored securely in your system keychain.

$ more-life setup
  Opening WHOOP authorization...
  Waiting for callback... authorized.
  Fetching 60 days of history... done.
  Calculating your baselines... done.

  ✓ Ready. Your AI now has a health instinct.

Step 3

Check your health

Run your first health check. This shows today's metrics compared against your personal baselines.

$ more-life health --pretty

  Today's check-in:

  Recovery is sitting around 68%, which is close to
  your usual 65%. Nothing out of the ordinary.

  HRV is 46ms — right in your normal range.
  Sleep was 7.2 hours with good efficiency.

  Everything looks steady. Nothing needs
  your attention right now.

Step 4

Understand the silence score

The silence score determines whether your AI should speak up. It evaluates four dimensions:

• Confidence — Is the pattern real? (single data point = low, 3+ day trend = high)
• Stakes — Does it matter today? (light day = low, big event tomorrow = high)
• Timing — Can you act on it? (3am = low, morning planning window = high)
• Urgency — Is time running out? (general awareness = low, decision window closing = high)

Score = Confidence × Stakes × Timing × Urgency. If the score is ≥ 0.6, your AI speaks up. Otherwise: silence. That's the feature.

$ more-life assess --json | jq -r '.score'
  0.38

# Score is below 0.6 — nothing needs your attention

Step 5

Integrate with your AI agent

More Life works standalone, but it's designed to run as a skill for your AI agent. Here's the pattern your agent should follow:

# Your AI agent runs this check periodically

$ score=$(more-life assess --json | jq -r '.score')
$ if [ $(echo "$score >= 0.6" | bc -l) -eq 1 ]; then
>   more-life health --json  # Safe to mention health
> fi

# Works with Clawdbot, Moltbot, Claude Code,
# or any AI agent that can run CLI tools

What's next

Let it run for a week. Your baselines will calibrate. Your AI will learn when to speak and when to stay quiet. Most days, you won't hear anything — and that's the point.