System Status
Uptime
99.9%+
Target uptime
15 min
Health-check interval
24/7
Automated canary testing
How We Monitor
-
Heartbeat probe every 15 minutes — a GitHub Actions workflow hits
/healtharound the clock and verifies the response includes all expected feature flags. - Daily paid canary — a separate workflow purchases a $0.001 tool once per day to verify the full end-to-end payment flow: tool discovery, x402 negotiation, USDC settlement on Base, and result delivery.
- Automated incident management — on any probe failure, a GitHub issue is auto-opened with diagnostic details. When the service recovers, the issue is auto-closed. No open issues = healthy.
- Transparent monitoring — all monitoring runs in the open. View the heartbeat workflow runs and open issues on GitHub.
Infrastructure
Railway
Managed hosting with zero-downtime deploys
Base L2
USDC settlement on Coinbase's L2 chain
Coinbase CDP
x402 payment facilitation
Global CDN
Edge caching for static assets
Live Endpoints
| Endpoint | Description |
|---|---|
| /healthFREE | Server health check with feature flags, version, and tool count |
| /api/reliabilityFREE | Uptime percentage, error rate, and response-time report |
| /api/statsFREE | Economy stats: total calls served, revenue, and tool-level breakdown |
Incident Response
Automated detection: When the heartbeat probe or paid canary fails, a GitHub issue is automatically opened with the failure details, HTTP status, and timestamp.
Maintainer notification: The repository owner is notified immediately via GitHub's issue notification system. Critical failures trigger immediate investigation.
Resolution: Most incidents are infrastructure-related (upstream API latency, Railway restarts) and resolve within minutes. The heartbeat re-probes every 15 minutes and auto-closes the issue on recovery.
Post-incident: Persistent issues are tracked in the GitHub issue tracker with root-cause analysis and remediation steps.