Health Monitoring

Health score

Every health check produces a health score — a number from 0 to 100 that represents the overall condition of your device. The score is broken down into five categories, each weighted and scored independently. Noah also assigns a letter grade for quick readability:

The overall score is a weighted composite of the five category scores. Each category contributes to the total based on its relative importance — security issues weigh more heavily than, say, a missing backup configuration.

Check categories

Noah evaluates your device across five categories. Each category runs a set of specific checks tailored to your operating system.

Security

Verifies that your device follows security best practices:

• Firewall — Is the system firewall enabled and configured?
• Disk encryption — Is FileVault (macOS), BitLocker (Windows), or LUKS (Linux) active?
• Screen lock — Is the screen set to lock automatically after a period of inactivity?

Updates

Checks whether the operating system and critical software are up to date:

• OS patches — Are there pending system updates?
• Update configuration — Are automatic updates enabled?

Backups

Confirms that a backup solution is configured and running:

• Time Machine (macOS) — Is Time Machine enabled? When was the last backup?
• File History (Windows) — Is File History or another backup solution configured?
• Backup recency — Has a backup completed within a reasonable timeframe?

Performance

Monitors system resources to detect slowdowns or bottlenecks:

• CPU usage — Is sustained CPU usage abnormally high?
• Memory pressure — Is the system running low on available memory?
• Disk space — Is the boot drive running out of free space?

Network

Tests connectivity and network health:

• Internet connectivity — Can the device reach the internet?
• DNS resolution — Is DNS resolving correctly?
• Latency — Is network latency within acceptable ranges?

Scanner system

Under the hood, health monitoring is powered by a set of scanners — specialized modules that each handle one category of checks. Noah ships with six scanner types:

Platform-specific behavior

Each scanner adapts its checks to the current operating system. The same scanner type may use different system tools depending on the platform:

On-demand scanning

You can trigger a health check at any time from the Noah app. Just click the health score indicator or ask Noah to "run a health check." Results appear within a few seconds.

Pause and resume

If you need uninterrupted system performance (for example, during a presentation or video call), you can pause health scanning. Noah will resume automatically when you unpause, or after a timeout.

Cooldowns

After a scan completes, each scanner enters a short cooldown period to avoid redundant rechecks. This prevents excessive resource usage if multiple triggers fire in quick succession.

Progress events

While a scan is running, Noah emits progress events to the UI. You'll see which scanner is currently active and how far along the overall check is. This keeps you informed without requiring you to wait for the full scan to finish.

Scheduling

Health checks run on a predictable schedule so your device stays monitored without any manual effort.

Automatic schedule

• Startup delay — After Noah launches, it waits 5 minutes before running the first health check. This gives the system time to finish its own startup tasks and avoids competing for resources.
• Recurring interval — After the initial check, scans repeat every 6 hours. This cadence is frequent enough to catch emerging issues but light enough to stay invisible.
• Idle awareness — Scanners respect the system's idle state. If you're actively using the machine, Noah may defer non-critical checks until the system is idle to avoid impacting your work.

On-demand checks

The automatic schedule doesn't prevent you from running checks manually. You can trigger a full scan at any time, and the next scheduled scan resets from that point.

Auto-heal integration

Health monitoring becomes more powerful when paired with Noah's auto-heal system. When a health check detects a failing condition, auto-heal can automatically trigger the right playbook to fix it — no user intervention required.

How it works

1. Detection — A health scanner identifies a failing check (e.g., "firewall is disabled" or "no backup in 30 days").
2. Triage — Noah's AI evaluates the failing checks and selects the best playbook to address each one.
3. Remediation — If auto-heal is enabled, Noah creates a new session, activates the selected playbook, and executes it autonomously. Actions that require elevated permissions or carry risk still prompt for your approval.
4. Cooldown — After an auto-heal attempt (whether it succeeds or fails), that specific check enters a 24-hour cooldown. Noah won't retry the same fix repeatedly.

Configuration

Auto-heal is off by default. You can enable it in Noah's settings. When disabled, Noah still detects issues and notifies you, but it won't take corrective action without your explicit request.

Connection to playbooks

Auto-heal relies entirely on the playbook system for its remediation logic. When a health check fails, Noah looks for a playbook that matches the issue — whether it's a bundled playbook, a local playbook you wrote, or a fleet playbook pushed by your admin. The same precedence rules apply: fleet playbooks override bundled ones, and bundled override local.

For a deeper look at how playbooks and auto-heal interact, see Playbooks: Auto-heal.

Proactive monitoring

Beyond scheduled health checks, Noah can run proactive background diagnostics that watch for emerging problems before they become critical.

What it monitors

• Disk usage trends — Noah tracks disk space over time and warns you if you're on track to run out within days, not just when you're already low.
• CPU hogs — Detects processes that are consuming excessive CPU over sustained periods. This catches runaway processes, stuck updates, and misbehaving apps.
• Crash logs — Scans system crash logs and application crash reports for patterns. If an app is crashing repeatedly, Noah surfaces it before you notice.

LLM triage

When proactive monitoring detects something noteworthy, Noah's AI triages the finding. It evaluates the severity, checks whether the issue is likely to resolve on its own, and generates a plain-language suggestion explaining what it found and what you can do about it.

Acting on suggestions

Suggestions appear as notifications in the Noah app. For each suggestion, you can:

• Act on it — Noah opens a session and guides you through the fix (or runs a playbook automatically if auto-heal is on).
• Dismiss it — If the issue isn't relevant or you've already handled it, dismiss the suggestion. Noah learns from dismissals to reduce noise over time.

Toggle on or off

Proactive monitoring can be toggled independently from health checks in Noah's settings. If you want the regular health score but don't need background diagnostics, you can turn proactive monitoring off while keeping scheduled health checks active.

Health reports

Noah can generate a comprehensive health report summarizing the current state of your device. Reports are useful for compliance audits, IT reviews, and keeping a record of your device's condition over time.

What's in a report

• Overall health score and letter grade
• Per-category scores with individual check results
• List of passing and failing checks with details
• Timestamp and device information
• Recommendations for any failing checks

Exporting

You can export a health report as a plain text file for easy sharing, email attachment, or archival. Ask Noah to "export a health report" or use the export option in the health screen.

Health history

All health check results are stored locally in a SQLite database on your device. This lets Noah track your health score over time and show you trends — is your score improving, holding steady, or declining? The history view helps you understand the long-term trajectory of your device's health.

Fleet sync

If your device is linked to a Noah Fleet dashboard, health scores sync to the admin dashboard automatically after every check. This gives your IT team visibility into the health of all managed devices without requiring manual reporting.

What gets synced

• Overall health score and letter grade
• Per-category scores
• List of failing checks (names and severity, not raw system data)
• Timestamp of each check
• Auto-heal run reports, if applicable

How it works

After each health check completes, Noah sends the results to the fleet server during its next checkin. Checkins happen every 6 hours (aligned with health checks) and on-demand when triggered by the fleet dashboard. The admin sees an aggregated view of all device scores, can drill into individual devices, and can push fleet playbooks to address common issues across the organization.

Privacy

Fleet sync shares health scores and check results, not raw system data. Your admin sees "firewall is disabled" but not the contents of your firewall configuration. Detailed system information stays on your device.