Monitor your APIs before your users notice.
High-frequency synthetic health checks, TLS certificate tracking on every probe, and multi-channel incident dispatch to Discord, Telegram & Email.
Connecting to the probe worker…
The console above is only populated once the background worker has written its first real snapshot.
How API Watch protects your uptime.
Register a public HTTP or HTTPS endpoint, pick an interval, and the worker starts probing on its next cycle. No agent to install, nothing to wire up.
Declare Endpoint
Public HTTP & HTTPS Targets
Specify any API destination and probe frequency. Loopback addresses, private RFC-1918 subnets, and malicious redirect attacks are blocked at ingress.
Continuous Probes
Single Background Worker
A dedicated scheduler measures DNS resolution, TLS handshake and total round-trip time, persisting every result in SQLite on the same instance that serves the app.
Multi-Channel Alerts
Discord, Telegram & Email
When an endpoint fails three consecutive probes or recovers, notifications go out with latency, HTTP code and the incident context. Discord and Telegram need a PRO plan; email goes through the Resend API.
Engineered for developer peace of mind.
Every claim on this page maps to code you can read. Here is the part of the engine that decides whether to wake you up.
High-Frequency Synthetic Probes
Millisecond-resolution timing
A single background worker runs scheduled probes and records status code, round-trip latency, DNS and TLS timings. Free checks every 15 minutes; Pro adds 1, 5, 10 and 15-minute intervals. Every plan can see p50/p90/p99 latency on the monitor detail page.
Alerts Without the Noise
Three strikes before an incident
Network jitter and one-off blips do not page anyone. The state machine needs three consecutive failed probes to open an incident, and a single successful probe resolves it and sends the recovery notice.
TLS Tracking & Encrypted Secrets
Certificate read on every probe
Each probe reads the served certificate and reports days remaining inside outage and recovery alerts, so an expiring cert shows up in the incident you are already reading. Monitor URLs, request headers and alert webhooks are stored with AES-256-GCM.
Isolated Public Status Pages
Private until you publish it
Each account gets its own status page slug. Monitors are private by default and endpoint URLs plus alert secrets are stripped before anything is serialized for the public page.
Maintenance Windows
Pause without losing history
A PRO-only window suspends probing for a planned deploy. Paused monitors are excluded from alerts and shown as intentionally paused rather than down, and the worker resumes them automatically when the window closes.
Three strikes before we wake you
A single failed probe is usually noise. An incident is declared only after three consecutive failures, and one success resolves it.
All probes passing. No open incident.
Timeline above is compressed for demo. In production the three strikes are consecutive scheduled probes: about 3 minutes apart on a 1-minute interval, 45 minutes apart on a 15-minute one.
Provision monitors from your deploy pipeline.
One POST to /api/v1/monitors with a Bearer API key. Create a key in Settings, then call it from curl, Python, Go or CI.
# Register a 1-minute probe with a keyword assertion.
# Intervals must match your plan: Free is 15, Pro accepts 1, 5, 10 and 15.
# Alert webhooks are set from the dashboard (PRO), not here.
curl -X POST "$API_WATCH_URL/api/v1/monitors" \
-H "Authorization: Bearer $API_WATCH_KEY" \
-H "Content-Type: application/json" \
-d '{
"name": "Payments API",
"url": "https://api.example.com/v1/health",
"method": "GET",
"interval": 1,
"expectedStatusCode": 200,
"keywordAssertion": "\"status\":\"ok\""
}'
# 201 Created -> { "monitor": { "id": "...", ... } }Set API_WATCH_URL to your instance origin and API_WATCH_KEY to a key from Settings.
Replace the example hosts with your own: the API resolves the hostname before creating the monitor and rejects anything that does not resolve, or that resolves to a private address.
Predictable plans for growing teams.
Start with 3 free synthetic monitors. Upgrade anytime for 1-minute intervals, multi-channel alerts and maintenance windows.
Free Developer
Essential uptime monitoring for hobbyists & side projects.
No credit card required. Up to 3 live endpoints.
- 3 Synthetic Monitors
- 15-minute check frequency
- 3-strike incident protection & anti-flapping
- Transactional email failure alerts
- Public status page & latency percentiles
- SLA audit report: uptime %, MTTR, p50/p90/p99, with JSON export
Pro Production
For production apps, mission-critical APIs & teams.
Billed monthly · Cancel anytime
- 25 Synthetic Monitors
- 1-minute high-frequency checks (also 5, 10, 15 minutes)
- Discord Webhooks & Telegram Bot alerts
- Maintenance windows — pause a planned deploy without losing history
- 30-day history with daily aggregates (Free keeps 1 day)
- Payload & keyword assertion verification (POST body + headers)
- Self-host the whole stack on your own VPS for $0/mo
Frequently asked questions.
Essential facts on probe architecture, notification vectors, and plan boundaries.