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

# Clients: every workspace on one page

> The org home at /clients: each workspace's loop status, work shipped, what waits on approval and visibility as counts, with bulk approval across workspaces.

If you run Attensira for several brands, each brand is a workspace in one org. **Clients** (at `/clients`, in the sidebar when the org has more than one workspace) is the org home: one card per workspace.

On the [Agency plan](/account/plans#agency) there is no limit on client workspaces. No plan counts workspaces; an org's prompt slots, responses and credits are shared across all of its workspaces.

## What each card shows

* **Loop status**: working, idle or blocked, with the reasons.
* **Shipped**: work shipped in the last 7 days.
* **Waiting**: items waiting on approval, with open tasks counted separately.
* **Visibility**: "named in X of N answers" over the last 7 days, summed across engines.

There is no week-over-week comparison on the page.

### Loop status on the card

For now the card's status is a summary derived from the workspace's automations:

| Status      | When                                                                         |
| ----------- | ---------------------------------------------------------------------------- |
| **Blocked** | No scheduled automation, all of them paused, or the credits are used up      |
| **Idle**    | An automation is scheduled but has never run, or nothing has run in 48 hours |
| **Working** | Otherwise                                                                    |

It will read [loop health](/agent/loop-health) instead once that is wired into the card; loop health is the fuller answer and is already on each workspace's Today page.

## Approving across workspaces

Select suggested changes on several cards and approve them in one action.

* **Select all** covers only the changes each card shows, up to 5 per card.
* A selection over 100 changes is sent in batches.
* Each change is approved exactly as it would be in that workspace's Work page: the same approval mode, change caps and checks apply.
* Failures are reported per item, so one refused change does not hide the others that went through.

## Client link

Each card has a **Client link** control that creates the [client reviewer link](/agencies/client-reviewer-link) for that workspace.

## Over the API

These are dashboard routes (signed-in session):

* `GET /v2/org/overview`: the cards.
* `POST /v2/org/inbox/approve` with `{"items": [{"project_id": "...", "item_id": "..."}]}`, at most 100 items. It answers `{approved, failed, results: [{project_id, item_id, ok, state?, error?}]}`.