Health Check API

Intent

Expose endpoints that report service liveness/readiness so orchestrators and operators can manage instances safely.

Use when

• Always in production services: orchestrators need readiness/liveness to route traffic correctly.
• You want to implement safe deploys and reduce blast radius during partial failures.

Avoid / watch-outs

• Don’t make health checks depend on every downstream dependency; distinguish “ready” vs “degraded”.
• Rate-limit and keep checks fast; health checks themselves shouldn’t cause load issues.

Skill mapping

• observability: define liveness/readiness semantics and implement consistent endpoints.
• resilience: decide how dependency health affects readiness and load shedding.
• platform: provide shared health check helpers across services.