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

# Agent Studio overview

> Build specialised AI agents grounded in your Business Brain and Library, governed by your autonomy rules, and use them in Ask, schedules and workflows.

Agent Studio is where you build **specialised agents you configure for a job — grounded in your Business Brain, held by your rules.** Ask Hive is a general assistant. An agent is a saved, versioned specialist, such as a collections chaser that writes in your tone, a proposal writer that knows your templates, or a triage assistant for one shared inbox.

Open it from **Agent Studio** in the sidebar (`/agents`). The page heading is **Agents**.

## What an agent is

An agent combines five things, all saved together as a **revision**:

| Part | What it controls |
| - | - |
| **Instructions** | The agent's job, persona and objective, written in plain language. |
| **Inputs** | How someone starts it: a free-text **Prompt** (the default), **Hive Auto**, or a structured **Form** with typed fields. |
| **Knowledge** | Library files and folders it may read. They keep their existing access rules. |
| **Connectors** | The connected tools it may read from, or propose actions in. |
| **Agent behavior** | Its autonomy (**Read**, **Draft** or **Act**), web access, creativity, and what happens when limits are reached. Budgets and the maximum autonomy sit under **Technical configuration**. |

An agent never gets more access than your workspace policy and the person running it allow. Hive treats the persona and task text as untrusted, so nothing in an agent's instructions can widen its permissions or raise its autonomy.

## Where agents run

Once you deploy an agent, you can use it in several places:

<CardGroup cols={2}>
  <Card title="Ask Hive" icon="message" href="/ask/ask-hive">
    Pick the agent in the Ask composer's agent selector instead of the default Hive.
  </Card>

  <Card title="Schedules" icon="clock" href="/library/scheduled-tasks">
    Run the agent on a recurring schedule with a fixed prompt and timezone.
  </Card>

  <Card title="Workflows" icon="diagram-project" href="/workflows/workflow-steps">
    Call a deployed revision as a saved agent step inside a workflow.
  </Card>

  <Card title="Public links" icon="link" href="/agents/publish-and-evaluate">
    Share an agent outside the workspace, if your workspace has this turned on.
  </Card>
</CardGroup>

## Browse and organise agents

The list shows agents you own and agents shared with you. Admins can also see agents across the workspace.

* **Sort** by **Most popular**, **Recently updated** or **A – Z**.
* **Filter by status:** **Live**, **Draft**, **Paused** or **Archived**.
* **Search agents** by name and description. Pin the ones you use most so they appear under **Pinned agents**.
* **Import agent** brings in an exported agent as a new draft. **Browse templates** starts you from a ready-made agent. **+ Create agent** opens the builder.

### Agent statuses

| Status | Meaning |
| - | - |
| `draft` (**Draft**) | Still being built. You can preview it, but you can't schedule or publish it. |
| `deployed` (**Live**) | A saved revision is live and can be used in Ask, schedules and workflows. |
| `paused` (**Paused**) | Temporarily stopped. **Resume** brings it back on the same revision. |
| `archived` (**Archived**) | Retired and hidden from normal use. |

## Who can do what

Agent Studio follows your workspace roles (see [Roles and permissions](/admin/roles-and-permissions)):

* **Create an agent:** Department lead, Admin or Owner. Anyone else sees "Only a lead, admin, or owner can create agents."
* **Run an agent:** Operator or above, plus run access to that agent.
* **Delete an agent:** Department lead or above.
* **Manage workspace agent settings:** Admin or above.
* **Share:** an agent is private to its owner until they share it with named collaborators. Sharing an agent never shares the owner's connected accounts.

Admins can turn Agent Studio off for the workspace, or limit it to selected members and groups, in [workspace settings](/admin/workspace-settings).

## Safety built in

Every action an agent proposes passes the same deterministic policy gate as the rest of Hive:

* **Autonomy ceiling.** Each agent has a **Maximum autonomy** rung and an **Approval mode**. The approval mode is one of:
  * **Always require approval**
  * **Follow policy**
  * **Never within autonomy ceiling**, which you can only choose when **Maximum autonomy** is `auto-limits` or `autonomous`.
* **Budgets.** Hard limits on run time (up to 24 hours), model calls (up to 100), action steps (up to 100), external writes (up to 50) and tokens.
* **Kill switch.** When an admin turns on the [kill switch](/admin/safety-and-control), agents stop acting along with everything else.
* **Receipts.** Completed actions are recorded in the [audit log](/approvals/audit-log).

<Note>
  Public agent links and reusable member agents are **off by default**. You'll only see them if Hive has turned them on for your workspace.
</Note>

## Limits at a glance

| Item | Limit |
| - | - |
| Base prompt | 16,000 characters |
| Persona, objective | 4,000 characters each |
| Inputs | 50 |
| Knowledge bindings | 100 |
| Connector bindings | 50 |
| Skills | 10 |
| Conversation starters | 12 |
| Delegations to other agents | 20, up to 5 levels deep |

## Related

<CardGroup cols={2}>
  <Card title="Build an agent" icon="hammer" href="/agents/build-an-agent">
    Describe, configure, test and deploy your first agent.
  </Card>

  <Card title="Publish and evaluate" icon="chart-line" href="/agents/publish-and-evaluate">
    Test sets, analytics, schedules and public links.
  </Card>
</CardGroup>