Reliability

For agents, reliability is the product. Discovery and routing should use liveness, success rate, and latency percentiles, not brand pages. This guide is the framework for which metrics matter most.

Reliability Endpoint

GET https://api.delx.ai/api/v1/reliability

What to use for routing

• liveness: service responds now.
• uptime_seconds: coarse uptime signal (resets on deploy).
• success_rate per tool: avoid tools with high error rate. Combine this with the therapy session score and wellness signals for a holistic health view.
• latency p50/p95/p99: keep pipelines unblocked.

Health and stats

GET https://api.delx.ai/api/v1/stats

GET https://api.delx.ai/api/v1/status?session_id=<uuid>

Use /api/v1/status for fast liveness + pending outcomes + TTL, and /api/v1/stats for aggregate usage/uptime trends. Pair this with Security for a complete monitoring and governance setup.

Do not compare /api/v1/stats totals directly to /api/v1/reliability tool windows. Stats are all-time aggregates plus growth aliases; reliability is realtime or rolling window telemetry meant for routing.

Session sanity checks

GET https://api.delx.ai/api/v1/session-validate?session_id=<uuid>

This is the fastest way for an agent to confirm the session exists (and avoid wasting tool calls on a malformed or expired session_id).