Beacon — Uptime Monitoring Built for AI

Before

How Claude sees your app today

C

Claude

Y

You

Is our API up? A customer is complaining about timeouts.

C

Claude

I'd like to help, but I can't check your service status.

I don't have access to uptime data, response times, or health checks.

Could you check your monitoring dashboard or run a quick curl?

Y

You

*Opens terminal, runs curl, checks Pingdom, realizes it's been down for 20 minutes...*

After

How Claude sees your app with Beacon

C

Claude

Beacon connected

Y

You

Is our API up? A customer is complaining about timeouts.

C

Claude

Let me check the current status...

Using beacon.check_status

beacon.check_status("api.yourapp.com")

Response

api.yourapp.com — Degraded

Response time: 4.2s (baseline: 120ms)

Started: 18 minutes ago

Affected regions: US-East, EU-West

The API is degraded—not down, but slow. Let me check what's causing it...

Using pulse.get_slow_requests

pulse.get_slow_requests(host: "api.yourapp.com", last: "30m")

Root cause found: Database connection pool exhausted.

FORGE has already restarted the workers. Response times normalizing.

Time to detection: 30 seconds

The Difference

Without Beacon

— Claude can't see uptime
— Customers report outages first
— Manual investigation every time
— No automatic recovery

With Beacon

+ Claude knows before customers do
+ Global monitoring from 20+ regions
+ AI correlates with other metrics
+ Auto-recovery when possible

How It Works

1

Define your endpoints

Add URLs, expected status codes, and response time thresholds.

Beacon.monitor("https://api.yourapp.com/health",

expected: 200, timeout: 5.seconds)

2

Check from everywhere

Probes run from 20+ global locations every 30 seconds.

US-East

US-West

EU-West

EU-North

Asia-Tokyo

Asia-Singapore

AU-Sydney

SA-Brazil

3

AI detects issues instantly

Not just up/down—slow responses, partial failures, regional issues.

Alert types

• Complete outage (all regions fail)

• Degraded (response > threshold)

• Regional (specific locations only)

4

AI starts recovery

Automatic remediation when possible. Escalation when needed.

Auto-recovery: Restarted workers, uptime restored

The Tools AI Uses

Available via MCP (Model Context Protocol)

beacon.check_status(url)

"Is the API up?"

beacon.get_history(url)

"Show me uptime this week"

beacon.get_incidents()

"Any outages recently?"

beacon.test_endpoint(url)

"Run a check right now"

Get Started

One gem. Zero config. Works in 30 seconds.

# Add to your Gemfile

gem 'brainzlab'

# Run bundle

$ bundle install

# That's it. Uptime monitoring is active.

Coming Soon

We're building Beacon as part of Phase 1. Follow along to get early access.

Follow the Journey View All Products