Skip to main content

Dashboard

What it is for​

The dashboard answers one question: are my agents working? Open it first thing and you should be able to tell in a glance whether last night went well.

It is the app's landing page, so it is what a new user sees before they know what Hermiq does.

What you see​

Four numbers across the top, computed from the run audit trail rather than a separate store:

  • Total runs, over every agent you can see
  • Success rate, the share of those runs that finished cleanly
  • Average latency, how long a run takes
  • Tokens, LLM usage where the provider reported it

Below them:

  • Running flows, anything in flight right now
  • Runs by agent, so a single misbehaving agent stands out from the rest
  • Schedules quota and Agents quota, against your organisation's limits
  • Agents, the most recent ones, as a table you can click through

What the numbers count​

Every figure is scoped to the agents you may see. A run belonging to another organisation is never counted, and the boundary is the agent rather than the schedule, because a flow-triggered run has no schedule.

Dry runs and replay previews are excluded. A preview must not move a real success rate.

Where the provider recorded no token usage, tokens read as unavailable rather than zero. A confident zero is worse than an honest gap.

API​

  • GET /apps/hermiq/api/analytics returns the metrics, optionally scoped with agentId

Where to go next​

The dashboard tells you that something went wrong. It does not tell you what. Open Runs and filter on the failed status.