Intelligent Agents helps you monitor an org unit to find activity that matches criteria you define. The criteria that agents search for are login activity, course activity, and release conditions in Brightspace.
Use Intelligent Agents to identify users and automatically take action when criteria are met. This helps you proactively support learners and automate repetitive tasks in your course.
Start with a goal
Before you create an intelligent agent, identify what you want to achieve.
For example, you might want to:
- Re-engage inactive learners using Login Activity or Course Activity
- Notify learners based on grades using Release Conditions
- Automate enrollment changes using Actions > Change Enrollments
- Track engagement patterns using Practice Run and View History
These examples show how a clear goal directly shapes how your agent is configured.
| Type |
Message |
|---|
|
Tip: If you cannot clearly define your outcome, your agent configuration will likely not behave as expected. |
Choose your next step
Use this page to understand how Intelligent Agents work. Refer to the following topics based on what you want to do next:
Example uses for agents include:
- Emailing users with grades below a certain level
- Checking for users who have not logged in within a specific number of days
- Checking for users who view a specific content topic
What Intelligent Agents do
Monitor user activity
Agents evaluate login activity, course activity, and a wide range of release conditions in Brightspace, such as assignment submissions, quiz attempts, grades, and discussion participation.
Trigger automated actions
When users meet the specified criteria, agents can perform actions automatically, including the following:
- Send automated email notifications to learners, instructors, or other stakeholders
- Change user enrollments by adding or removing users from courses or groups
- Change user roles
- Support awards or certificates through release conditions
Support engagement and retention
Common use cases include:
- Re-engaging inactive learners, for example, users who have not logged in for seven days
- Notifying learners or instructors about low grades or missed activities
- Recognizing achievements, such as completion of a module
- Automating administrative workflows, such as unenrolling users after course completion
Automate repetitive tasks
Agents reduce manual effort by automating routine communications and course management actions.
Requirements
The following are required to use Intelligent Agents:
- To see and use Intelligent Agents, you must have the required Intelligent Agents permissions applied to your role. These permissions can only be accessed by your organization administrator.
- To locate Intelligent Agents, navigate to Course Admin from your navbar.
|
Important: If you cannot find Intelligent Agents in your Brightspace instance, contact your organization administrator to obtain required permissions. |
How Intelligent Agents work
Intelligent Agents execute actions based on defined criteria at the time of execution.
Intelligent Agents follow this process:
- You define the criteria for the agent based on user roles in the classlist, activity data, and release conditions.
- You select whether you want to schedule the agent to run automatically or trigger it to run manually.
- When the agent runs, it evaluates which users currently meet the criteria.
- If users meet the criteria, the agent performs the configured action.
- Results are logged for auditing and troubleshooting purposes.
|
Important: Intelligent Agents do not continuously monitor user activity. Evaluation occurs only at execution time, using each user’s current state. |
To validate results:
- Practice Run: Executes the evaluation logic without performing actions, allowing validation of criteria and expected outcomes.
- Run Now: Immediately executes both evaluation and actions for all users who meet the criteria at that time.
Visual tour of Intelligent Agents
- Create a new agent, create and edit categories to group related agents together, and select More Actions for more options.
- Enable, Disable, Delete, or Bulk Edit existing agents.
- Use the table to select agents, edit agents, and review information about agent run dates.
- Use the drop-down menu to filter agents.
- Customize email settings related to Intelligent Agents.
Watch how to encourage engagement using Intelligent Agents
Considerations and best practices for Intelligent Agents
When you set up an intelligent agent, you must decide on the following:
What the agent is looking for
The agent looks for specific conditions or criteria, such as:
- User login activity
- Course activity
- Release conditions (rules like completing or not completing something)
|
Important: Agents check these conditions at the moment they run. This means a user who met the condition earlier may not meet it later, especially for “NOT” conditions (for example, has NOT logged in). |
Understand criteria behavior
- Criteria are evaluated only at runtime.
- Conditions that use negative logic can change over time.
- A user can meet the criteria during one run and not meet them during a later run.
How often the agent will run
You decide how often the agent checks for users who meet the criteria. You can choose to set a run schedule or to run the agent manually:
- If you set a schedule, it runs automatically.
- If you don’t set a schedule, you must run it manually from the Agent List page.
What action the agent takes
When the criteria are met, the agent can:
- Change course enrollment (in the current course or another course)
- Send an email to the user, their auditor, or another person
- Take no action (useful if you only want reports showing who meets the criteria)
Replacement strings and language locale settings
Your agent can use replacement strings to automatically personalize a learner's last course access date or last login date in an email. If you decide to use replacement strings, you must consider your organization's default system locale language settings and your course offering level locale settings.
For example, if your organization is set to a default system locale of English (United States), but the course that your intelligent agent is sent from is set to Override locale preferences to a different language, the replacement string text will use the course offering locale settings.
Specifically, the following replacement strings are translated based on either the default locale settings for your organization or the override locale you set at the course offering level:
- {OrgUnitStartDate} - The start date specified for the org unit.
- {OrgUnitEndDate} - The end date specified for the org unit.
- {LastCourseAccessDate} - The date the initiating user last accessed the course.
- {LastLoginDate} - The date the initiating user last logged in.
|
Note: For more information about setting your default locale, refer to Set a default locale. To learn more about the replacement strings you can use in an intelligent agent, refer to the Replace strings available for intelligent agent emails section of Set up Intelligent Agents. |
How often the agent takes action
You choose whether the agent should:
- Act once when a user first meets the criteria, or
- Act every time it runs (for example, sending reminder emails).
Understand repetition
The Repetition setting controls how often all selected actions are applied, not only email actions.
- Take action only the first time the agent’s criteria are satisfied for a user processes a user once.
- Take action every time the agent is evaluated and the agent’s criteria are satisfied for a user processes the user every time the agent runs and the criteria are met.
Example
Criteria: Grade greater than or equal to 0%
Action: Change role
Repetition: First time
In this case, the role changes once. If someone manually changes the role back later, the agent does not run again for that user.
Common scenarios
1. Re-engage inactive learners
Goal: Email learners who have not accessed the course in 7 days.
Configuration:
- Role = Learner
- Login Activity: Not accessed course for 7 days
- Action: Send email
- Repetition: Take action every time the agent’s criteria are satisfied for a user
Validation:
- Run a Practice Run
- Confirm that the correct users appear
Expected outcome: Only inactive learners receive emails. Active learners are excluded.
For setup steps, refer to Set up Intelligent Agents.
2. Identify at-risk learners
Goal: Identify learners scoring below 50%.
Configuration:
- Role = Learner
- Release Condition: Grade < 50%
- Action: Send email or notify the instructor
Validation:
- Run a Practice Run
- Review the returned user list and adjust the condition if needed
For setup steps, refer to Set up Intelligent Agents.
Best practices for using agents
To get the most value from intelligent agents, consider these tips:
- Use a consistent naming convention so the purpose of each agent is clear.
- Use the Description field to add notes or reminders about what needs to be updated each term.
- Decide if the agent should repeat (for example, sending reminders regularly).
- Use replacement strings to automatically personalize emails, such as inserting a learner’s name.
- Avoid creating too many agents, especially if the information can be shared with students in another way.
- Start with a clear goal.
- Test your logic with a Practice Run.
- Use repetition settings carefully.
- Review results before using broad automation.
Related articles