> For the complete documentation index, see [llms.txt](https://docs.harmony.io/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.harmony.io/ai-agents/getting-started/configuring-ai-agents.md).

# Configuring AI Agents

Configure when agents run, how approvals work, and provisioning behavior for each agent type.

## Understanding Agent Configuration

Each agent type has a configuration panel on the right side of the detail page. Options vary by agent but typically include:

* **Schedule** - When the agent runs (daily, weekly, etc.)
* **Approval** - Who must approve and how (for agents that require approval)
* **Provisioning** - Access duration, auto-approval (for app access and similar agents)
* **Agent-specific options** - Thresholds, ticket desk, filters, etc.

## Configuring Agent Settings

1. Open the agent detail page.
2. Use the configuration panel on the right.
3. Change options (schedule, approval, thresholds, etc.).
4. Click **Save** to apply.

![Agent detail page with configuration panel](/files/bc3w2yeo8LYdxSHBhprG)

## Setting Agent Schedules

In the **Run Schedule** section:

**Recurring schedules**

* **Frequency** - Daily, Weekly, Monthly, Quarterly (every 3 months), Semi-annually (every 6 months), Annually
* **Day** - For weekly: day of week, for monthly/quarterly/semi-annually/annually: day of month (e.g., first, last, specific day)
* **Time** - Hour and minute
* **Timezone** - When the schedule runs in local time

**Cron-based schedules**

Agents use cron expressions internally. The UI exposes frequency, day, time, and timezone, these are converted to a 5-field cron expression for the backend.

**One-time runs**

Agents can be run manually. Use the **Run** button (or Run dialog) to trigger a one-time execution. This does not affect the schedule.

![Run Schedule section (frequency, day, time, timezone)](/files/HI9gauOWrQJouMv3lq8M)

## Configuring Agent Approval Workflows

For agents that require approval (e.g., Application Access Request, Group Membership Management, Employee Account Termination):

**Approval strategy**

| Strategy                                       | Description                                                                    |
| ---------------------------------------------- | ------------------------------------------------------------------------------ |
| **Require approval from all**                  | Every listed approver must approve.                                            |
| **Require approval from one of**               | Any one approver can approve.                                                  |
| **Escalate if first approver doesn't respond** | First approver is tried, if no response within the timeout, escalation occurs. |

**Approvers**

* **Role-based** - Direct manager, app owners, IT approvers, budget approvers, etc. (options vary by agent)
* **Employee-based** - Specific people from the employee directory

Add up to 10 approvers. Configure per agent based on available role options.

**Timeout**

Set how many days (1–7) to wait for approval before timeout or escalation.

**Reminders**

Enable reminders (first and second) at configurable intervals (e.g., 4 hours, 24 hours) to nudge approvers.

![Approval strategy and approvers](/files/3WTATBoKkuFmke32SDkz)


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://docs.harmony.io/ai-agents/getting-started/configuring-ai-agents.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.