The problem nobody talks about
Your API returns 200 OK.
Your monitor says healthy.
Your users are seeing broken data.
This is a silent failure — and it's more common than you think.
// HTTP 200 OK — but...
{
"status": "ok",
"data": null,
"error": "DB_CONNECTION_FAILED"
}
Standard monitors check the status code and stop there.
They have no idea what's actually inside the response.
What I built
CheckAPI — API monitoring that validates the response body, not just the status code.
Four layers of detection:
1. Keyword validation
A string that must be present or absent in the response body.
- Present:
"status":"ok"must exist - Absent:
"error"must not exist
2. Regex pattern matching
"status":\s*"(ok|healthy)"
"balance":\s*[1-9]\d*
3. JSONPath assertions
Up to 10 per monitor, AND/OR logic, 11 operators.
$.status equals "healthy"
AND $.database equals "connected"
AND $.errorCount less than 1
4. Header assertions
Validate Content-Type, X-Status, or any custom header.
Both Regex and JSONPath have live testers built into the dashboard.
Test your pattern against a real response before saving.
Everything that's live today
- All HTTP methods (GET/POST/PUT/DELETE/PATCH/HEAD/OPTIONS)
- Custom headers + request body
- Heartbeat/Cron monitoring (64-char token + grace period)
- Public Status Page (
/status/{id}) — zero auth required - 5 alert channels: Email, Slack, Telegram, Discord, Webhook
- Response time percentiles (p50/p95/p99)
- SLA reports (Pro/Business)
- Maintenance windows
- REST API access (Business plan)
Pricing
| Plan | Price | Monitors | Interval | History |
|---|---|---|---|---|
| Free | $0 | 10 | 5 min | 30 days |
| Starter | $5/mo | 20 | 1 min | 30 days |
| Pro | $15/mo | 100 | 30 sec | 90 days |
| Business | $49/mo | Unlimited | 10 sec | 365 days |
Free plan: commercial use allowed. Forever. No credit card required.
Honest build in public numbers
- Real users: 7
- MRR: $0 (payments are live)
- Solo founder from Seoul 🇰🇷
- Stack: FastAPI + Celery + Redis + PostgreSQL → Railway / Next.js 14 → Vercel
- Open source (MIT): github.com/JEONSEWON/CheckAPI
What's still missing
- Team UI (API already done)
- Multi-location checks
- TCP/Port/DNS monitors
- SLA Report PDF export
What I've learned
Getting a strong response to launch content is not the same as
validating that people will pay for your product.
I've had good engagement, real conversations, and zero conversions.
The hardest part isn't building the detection logic.
It's reaching developers who haven't been burned yet —
because they don't go looking for monitoring tools until something breaks.
Still figuring it out in public.
→ **[checkapi.io]
(https://www.checkapi.io/)
Honest feedback beats silence every time.
Top comments (0)