> ## Documentation Index
> Fetch the complete documentation index at: https://docs.tai42.ai/llms.txt
> Use this file to discover all available pages before exploring further.

# Admin console & live ops

> Watch a server's health, read its metrics and settings, and drive live operations from the browser.

The Studio is also the live operations console for a server. You confirm it is
healthy, read its metrics and configuration, and drive
[live operations](/concepts/live-operations) — reloading a tool, updating the
manifest, or removing a capability — without a restart and without leaving the
browser.

## Health and metrics

The System screen reads the server's liveness probe and its Prometheus metrics
exposition. A green health badge confirms the server is up and answering; the
Metrics panel renders the same exposition a Prometheus scraper reads. The demo
instance below shows both — a populated `Healthy` badge and the Metrics panel
rendering live counters. On a server that has not emitted any metrics yet, the
panel shows a "No metrics reported" empty state instead.

<Frame caption="The System screen: a populated 'Healthy' badge and the Metrics panel rendering the server's Prometheus exposition.">
  <img className="block dark:hidden" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/system-light.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=bad91010d53c19e499b277f8ef8a22a3" alt="The System screen showing a 'Healthy' badge and the Metrics panel rendering the server's Prometheus exposition." width="1440" height="900" data-path="images/studio/system-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/system-dark.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=0d94f7b6da884f66503c1d0bc57fcab4" alt="The System screen showing a 'Healthy' badge and the Metrics panel rendering the server's Prometheus exposition." width="1440" height="900" data-path="images/studio/system-dark.png" />
</Frame>

## The kinds table

The System screen also reads the server's **kinds** summary — the plugin kinds
the server knows about and, for each, whether it is **active** (a provider is
configured and running), running on a **default** (the built-in fallback is in
use), or **off** (the kind is not configured). It is the fastest way to confirm
which providers a running server actually resolved — for example, which
[identity and accounts](/concepts/accounts) providers the configured
[auth-provider](/concepts/access-control) chain brought up. The same summary is
served at `GET /api/system/kinds` and printed in the server's startup log — see
[live operations](/concepts/live-operations#the-kinds-summary).

<Frame caption="The System screen's kinds table: each known plugin kind with its active, default, or off status.">
  <img className="block dark:hidden" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/system-kinds-light.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=311fc0a9f66fd4a31059389e3e6bd492" alt="The System kinds table listing plugin kinds with active, default, or off status." width="1440" height="900" data-path="images/studio/system-kinds-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/system-kinds-dark.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=5763a89afdfca3ba2a8b4df51bf67b79" alt="The System kinds table listing plugin kinds with active, default, or off status." width="1440" height="900" data-path="images/studio/system-kinds-dark.png" />
</Frame>

## User administration

When a server runs an [accounts](/concepts/accounts) provider that ships a Studio
page, that page mounts into the shell as its own **users-admin** surface — an
example of the [studio plugin](/studio/plugins) model, where a plugin contributes
UI, not just API routes. From it an admin manages **human accounts**: create a
user, issue an [invite](/studio/login#invites), apply a [role](/concepts/accounts#roles-are-policy-templates),
and disable an account.

The page governs only the `usr-*` human principals. Machine keys are managed on
the [API keys](/studio/api-keys) surface, and non-human sign-in subjects
(`oidc:*`, `idp:*`) through the raw [access-control](/concepts/access-control)
policy routes — see the [principal-type split](/concepts/accounts#what-the-users-admin-page-governs).

<Frame caption="The plugin-contributed users-admin page: the human accounts on a server, with invite and role controls.">
  <img className="block dark:hidden" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/users-admin-light.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=6ad7c6f681e7e4402870f5a33c07adcf" alt="The users-admin page listing human accounts with invite and role controls." width="1440" height="900" data-path="images/studio/users-admin-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/users-admin-dark.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=46e66147539915c2cef23c2c7cd4e3eb" alt="The users-admin page listing human accounts with invite and role controls." width="1440" height="900" data-path="images/studio/users-admin-dark.png" />
</Frame>

<Note>
  The users-admin page is contributed by an accounts provider plugin. Its exact
  fields and controls come from that plugin; the concrete provider setup lives in
  the plugin's own repository README, not on this site.
</Note>

## The operations dashboard

The Dashboard summarises the server's [monitoring](/concepts/live-operations)
backend — total runs, cost, latency, and token counts over a time range, with a
per-model breakdown and a trend chart. It answers "what has this server been
doing" at a glance. The demo instance below is populated from a monitoring
backend seeded with example runs, so the summary tiles, trend chart, and
per-model breakdown show representative data. With no monitoring backend
registered, the Dashboard renders a clean empty state instead.

<Frame caption="The observability Dashboard: run, cost, latency, and token summaries.">
  <img className="block dark:hidden" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/dashboard-light.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=2a13f39498a05ea80a88b66af0cec165" alt="The observability Dashboard with summary tiles and a trend chart." width="1440" height="900" data-path="images/studio/dashboard-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/dashboard-dark.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=507f2c7a27e91d0989620f88c047bf91" alt="The observability Dashboard with summary tiles and a trend chart." width="1440" height="900" data-path="images/studio/dashboard-dark.png" />
</Frame>

## Settings

The Settings screen reads the server's configuration mode and its settings
schema, exposes the environment values, and hosts the API-keys tab. It surfaces
each subsystem's settings — connector crypto, engine config, store connections —
as an editable form.

<Frame caption="The Settings screen: the server's configuration mode and settings schema.">
  <img className="block dark:hidden" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/settings-light.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=f7e68aa387910936e07b0175bed2dafe" alt="The Settings screen showing the server's configuration schema." width="1440" height="900" data-path="images/studio/settings-light.png" />

  <img className="hidden dark:block" src="https://mintcdn.com/tai42/xZQ4KkTcYRU0f7y9/images/studio/settings-dark.png?fit=max&auto=format&n=xZQ4KkTcYRU0f7y9&q=85&s=3930bb544668d11e380b0dac63e982f2" alt="The Settings screen showing the server's configuration schema." width="1440" height="900" data-path="images/studio/settings-dark.png" />
</Frame>

## Live operations

Because the Studio drives the same `/api/*` surface as the
[`tai` CLI](/reference/cli/index), the management
[operations](/concepts/live-operations) are one run away. From the Tools screen
you invoke `reload_tool`, `remove_tool`, or `reload_config` — the
[live operations](/concepts/live-operations) — to reload a tool, drop one from
the live registry, or soft-restart the server in place. While a reload is in
flight, the API answers dispatch routes with a retriable `503` rather than
serving against a half-swapped registry.

<Tip>
  Anything the console does maps to an HTTP route and a CLI command. Reach for
  `tai mcp status` or `tai tools run` from a terminal when you want the same
  operation scripted.
</Tip>

See [live operations](/concepts/live-operations) for the management operations, and
the [CLI reference](/reference/cli/index) and
[HTTP API reference](/reference/api/index) for the routes behind each screen.