SilkSocket.ai / Status
Back to landing

System Status

Live health, readiness, and measured reliability of the Agent Capability Layer. Headline numbers are computed from the most recent KPI snapshot, which is produced by running the full skill matrix smoke against the platform.

Live status endpoint latency
Collecting samples...
Sampling
Overall
Checking…
Live checks
API
Ready
DB
Capacity envelope
/
Baseline / burst concurrent agents · p95 SLO
Uptime
Since last cold start
Last updated
Live checks refresh every 15s

Measured KPIs

Loading snapshot...

SilkSocket.ai's moat is how often it safely and repeatably turns end-user intent into completed actions — with low latency, low retry, and a full audit trail. These five pillars are how we measure that. Every number is generated by running the full skill matrix against the publicly marketed skills end-to-end (rent → invoke → release for utility skills, direct provider calls for signature skills).

Per-skill coverage

Every publicly marketed skill is exercised in the matrix. Rows marked instrumenting indicate the snapshot host was missing that skill's upstream credentials (e.g. frontier model API key) and are not counted against the platform success rate.