# How to setup custom escalation rule

#### Video Guide:

{% embed url="<https://player.mux.com/dnfLhpO7YaWi6NJVBYex8qUwouEUsOINNKwGnwdHzjM>" %}

#### Text Guide:

From your Dashboard:

1. Go to **Escalations**.
2. Under AI Safety Rules, click on **Create custom rule**.

<figure><img src="/files/aoS5IvPOSQA4Wh5Zt4C8" alt=""><figcaption></figcaption></figure>

**Rule name**

Give your rule a clear, descriptive name (for example, "Disable AI on chargeback threats") so your team can easily identify its purpose.

\
**When this rule matches**

Decide exactly what the AI should do when the rule's conditions are met.&#x20;

You have two main choices:

* **Disable AI:** This mutes the AI for the conversation. The AI stops replying, allowing a human agent to take over the thread manually.
* **Send Template:** Instead of the AI generating its own reply, it will draft a specific saved template (which you can create in your Templates settings). This draft will still flow through your standard approval queue.

**When should this rule trigger?**

Define the exact criteria that will cause this rule to fire. You can add one or multiple conditions to be as broad or specific as you need.&#x20;

<details>

<summary><strong>List of criteria you can base your triggers</strong></summary>

* Customer language
* Customer message (e.g., if the message contains specific words like "chargeback", "dispute", or "lawyer")
* Customer mood
* Order value
* Customer email
* Order status
* Number of products
* Product name
* Customer country
* Payment method
* Question type

</details>


---

# Agent Instructions: 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:

```
GET https://docs.scaleflows.net/documentation/documentation-v1.1/workspace/configuring-your-ai-automation/decision-configurations/escalations/how-to-setup-custom-escalation-rule.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
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.