Platform SLA
Service-level transparency your team can trust. System-measured — AI cannot declare or change SLA.
Last published snapshot: 2026-10-18 · served by GET /api/v22/platform/sla.
Live verification runs in your browser — no data about you is sent.
Overall
100/100 · 5/5 metrics healthy
| Metric | Value | Status | Healthy threshold |
|---|---|---|---|
| API availability | 99.9% | healthy | ≥ 99.5% |
| Data freshness | 92/100 | healthy | ≥ 85 |
| Source health | 95/100 | healthy | ≥ 90 |
| Processing latency | 180 ms | healthy | ≤ 300 ms |
| Review latency | 240 ms | healthy | ≤ 360 ms |
Metric definitions & health thresholds
- API availability — share of successful API requests; healthy at ≥ 99.5%.
- Data freshness — freshness score of synced intelligence sources (0–100); healthy at ≥ 85.
- Source health — share of configured sources syncing within their expected window (0–100); healthy at ≥ 90.
- Processing latency — pipeline processing time per intelligence signal; healthy at ≤ 300 ms.
- Review latency — time for human review of AI-drafted items; healthy at ≤ 360 ms.
This page publishes a service-level snapshot for transparency — it is not a contractual legal SLA. Thresholds are system-computed; AI can never declare, inflate or change them.
API: GET /api/v22/platform/sla. Related: Status · Platform Health · Observability