Health
GET /api/v1/health — unauthenticated liveness probe returning service status and version.
GET /api/v1/health is the unauthenticated liveness probe. Use it from load balancers, uptime monitors, and credential-test screens.
Request
GET /api/v1/health HTTP/1.1
Host: privacyflow.appNo headers required, no query parameters supported.
curl
curl -s https://privacyflow.app/api/v1/healthResponse
200 OK:
{
"status": "ok",
"version": "2.0.0"
}Schema
| Field | Type | Notes |
|---|---|---|
status | "ok" | Always "ok" when the service is alive. |
version | string | Semver package version. Bumps on breaking queue/body/env changes; the URL prefix (/api/v1) does not move in lock step. |
The response includes the standard security headers (X-Content-Type-Options, X-Frame-Options, Strict-Transport-Security, X-XSS-Protection) but no rate-limit headers — health is uncapped.
Status codes
| Code | When |
|---|---|
200 | Happy path. |
503 | Upstream Redis/DenoKV is unreachable. The service is up but degraded; treat as failing your health check. |
When to call it
- Uptime monitoring. Poll every 10–60 seconds from your monitoring platform. No auth required so it works behind any healthcheck service.
- Credential-test preflight. The n8n credential test and the MCP server’s
privacyflow_check_healthtool both call health before verifying the key — it gives the user a fast “is the API even reachable?” signal before burning an authenticated call. - Deployment validation. Hit
/api/v1/healthafter a deploy and assertstatus === "ok"before routing traffic.
Tip
Health returns the version. If you depend on a specific response shape (e.g., the messages array wrapper), assert the version too — your client can warn on a major version mismatch without crashing.
Last updated: