Helious status
All systems operational
As of 2026-10-25 21:47 UTC. Every surface is checked once a minute from outside, the way a customer reaches it. 100.000% uptime across all systems over the last 90 days.
Systems
Terminal
The web terminal at helious.io, served from the edge.
Operational
90 days ago
100.000% uptime over 90 days · target 99.9%
Today
Last checked 2026-10-25 21:46 UTC · 286 ms · HTTP 200 in 286 ms
REST API
api.helious.io/v1: rates, curve, calendar, auctions, series and the feed.
Operational
90 days ago
100.000% uptime over 90 days · target 99.9%
Today
Last checked 2026-10-25 21:46 UTC · 136 ms · HTTP 200 in 136 ms
Live stream
The WebSocket that pushes the tape, quotes and alerts to every open terminal.
Operational
90 days ago
100.000% uptime over 90 days · target 99.9%
Today
Last checked 2026-10-25 21:46 UTC · 2102 ms · init frame in 2102 ms
MCP server
The streamable-HTTP MCP endpoint that AI agents call.
Operational
90 days ago
100.000% uptime over 90 days · target 99.9%
Today
Last checked 2026-10-25 21:46 UTC · 72 ms · 25 tools in 72 ms
Market data and news pipeline
Release capture, the calendar, quotes and the headline feed behind every surface.
Operational
90 days ago
100.000% uptime over 90 days · target 99.5%
Today
Last checked 2026-10-25 21:46 UTC · 0 ms · light ok, feed age 6 s
Ask Helious
The answer engine's model provider, reachable and accepting our credentials.
Operational
90 days ago
100.000% uptime over 90 days · target 99.5%
Today
Last checked 2026-10-25 21:46 UTC · 517 ms · model provider answered in 517 ms
Billing and renewals
Checkout, subscription renewals and credit top-ups.
Operational
90 days ago
100.000% uptime over 90 days · target 99.5%
Today
Last checked 2026-10-25 21:46 UTC · 67 ms · heartbeat 7 min old
Past incidents
No incidents in the last 90 days.
How this is measured
A monitor on the Helious worker probes each surface once a minute over the public
internet: it loads the terminal, calls the REST API, opens the live stream and waits for the first
frame, and asks the MCP server for its tool list. The data pipeline is read from the same health
light the hosting platform uses to route traffic, Ask Helious from a credentialed call to the
model provider, and billing from the renewal biller's heartbeat.
A minute counts as down when the probe fails or does not answer in 10 seconds,
and as degraded when it answers slowly or partially. Uptime is the minutes that were not down
over the minutes checked, for the last 90 days. A day on the bar is painted by its worst
stretch: red from 5 minutes down, amber from 1. Three failed checks in a row open an incident
here and page the desk; three clean checks close it. Incidents the probes cannot see are opened
by hand and say so.
The targets are 99.9% for the terminal, REST API, live stream and MCP server,
and 99.5% for the pipeline, Ask Helious and billing. Machine readers can poll
https://api.helious.io/api/status, the same
data this page renders.
