Hi, Humans. How may we help you?

Conditions for executing the support specialist agent
5 min
Created by Leonora Alves on 7/14/2025 3:22 PM
Updated by Leonora Alves on 8/7/2025 12:36 PM

Conditions are the criteria that determine when a support expert agent will be executed. They are essential to ensure that the configured actions only occur in specific and relevant contexts within a ticket.

How condition blocks work

When setting up a support expert agent, you will define activation conditions within two blocks:

  • Combined Condition (AND): ALL conditions in this block must be true for the agent to be executed.
  • Independent Condition (OR): At least ONE of the conditions must be true for the agent to be executed.


If you use both blocks at the same time, the agent will only be executed if all conditions in the first block are met and at least one condition in the second block is true.

What are execution conditions?

Not every condition can execute an agent on its own. For the support expert to function properly, you must include at least one execution condition - that is, a condition that represents a specific event or change in the ticket.

See the difference:

Category

Example

Is it a execution condition?

Reason

Ticket status

Ticket: is → Open

✅ Yes

Represents a specific moment: when the ticket is opened.

Ticket: Status → Equals Closed

❌ No

It is only a current characteristic of the ticket, without indicating a moment of change.

⚠️ Attention: If your agent doesn't have any execution condition, it won't be activated.

You can use more than one execution condition, but remember to combine them carefully so they don’t prevent the agent from being executed.

Sentences as execution conditions

If the execution condition involves a sentence, the agent will only be executed with an exact match. For example: if the condition is “password change,” the agent will only be executed when that full phrase appears. Isolated terms like just “change” won’t be enough.

Time-based conditions

Time-based conditions (such as Dwell time or Not recorded for [X] hours) work best when combined with classification conditions, such as:

  • Urgency
  • Category
  • Service
  • Status
  • Justification

This combination improves performance and avoids unnecessary executions.

In summary:

  • Execution conditions: Execute the agent at the exact moment the condition occurs. Example: “Ticket: is Open” executes the agent as soon as the ticket is created.
  • Non-execution conditions: Act only as filters, checking whether the ticket meets specific criteria (like status or urgency). They don’t execute the agent on their own, but can be combined with execution conditions to control when the agent will run.


Practical use cases

See below some examples of conditions and how they can be used in practice:

Condition

How it works in practice

Ticket: is → Reopened

Execute the agent whenever a reopened ticket by a customer is identified. Can be used to automatically reclassify or reassign the ticket.

Ticket: Status → Equals to On Hold

Used to apply actions to tickets that are on hold, such as sending a notification to resume service or reassigning to another agent.

Ticket: Due date → Changed

Performs actions whenever the ticket's due date is changed. Ideal for logging the reason for the change or notifying the requester.

Ticket: Requester: Type → Equals to Company

Allows you to apply actions only when the requester is a company, such as automatically redirecting to a corporate support group.