Configure Copilot for Agents for Email Autosummary
This feature supports both Copilot for Agents and Copilot Automated Summary (AutoSummary). The term Copilot for Agents is sometimes used to refer to both unless noted otherwise.
You can use NiCE CXone GenAI Prompts to control the prompt that Copilot for Agents uses to create an autosummary for an email interaction. This lets you control the length, tone, and instructions of the autosummary.
Email AutoSummary generates a short summary of the email interaction after the interaction ends. The summary helps agents quickly capture the important details and outcome of the interaction.
Once you have completed your prompt configuration, you can publish the prompt to make it available for selection within the Copilot for Agents profile on the Copilot for Agents Config page.
All users can now open the LLM Model dropdown. Users with an Prompt Editor Advanced license can select any model in the list; other users are limited to standard models. The list itself is grouped by standard and premium models, with models verified by NiCE CXone shown first and the rest listed alphabetically.
NiCE Default is available as an LLM Model option. It represents the model NiCE CXone currently recommends for that specific prompt type. If a prompt template only supports one model, NiCE Default is the only option shown and the dropdown is disabled. If a template supports multiple models, NiCE Default appears alongside them as a selectable choice.
If the LLM model currently selected for a profile is scheduled for retirement, NiCE CXone displays a warning next to the LLM Model dropdown stating that the model will retire in a set number of days and prompting you to update to a supported model.
When a model you use is closer to its retirement date, NiCE CXone also displays an LLM Update dialog when you open the affected profile. The dialog states the date the model switches to retention mode, after which the model is no longer accessible, and reminds you to update the LLM and test and publish all affected profiles before that date. Select Don't show this message again to suppress the dialog for that profile, or click OK to close it.
If a model reaches its retirement date and you have not selected a different model, NiCE CXone automatically switches the profile to NiCE Default. Test and publish the profile after this switch to confirm the output still meets your requirements.
This table shows the LLMs you can choose from, along with a rating from 1-10 for the speed, accuracy, complexity handling, and hallucination resistance. 1 is always the lowest and 10 is always the highest score. A high score is desirable for speed, accuracy, and complexity handling, but not for cost and hallucination resistance.
-
This table lists only the LLM models that NiCE CXone has verified for this use case. The LLM Model dropdown in the application may show additional models that have not yet been verified by NiCE CXone; in the dropdown, an
icon next to a model indicates it has been verified by NiCE CXone. You can select any listed model, but if you choose one that isn't in this table, testing that model for your specific use case is your responsibility. -
NiCE Default is not a fixed model — it points to whichever model NiCE CXone currently recommends for this prompt type, and that underlying model may change over time. For this reason, NiCE Default isn't listed as a separate row in the table below.
-
Additionally, profiles previously defaulting to Nova Pro may now display NiCE Default instead, with no functional change.
| LLM Model | Speed | Accuracy | Complexity Handling | Hallucination resistance | notes |
|---|---|---|---|---|---|
| Nova Micro | 10 | 6 | 5 | 6 | Lightweight, low latency; limited reasoning. |
|
Nova Lite |
9 |
7 | 6 | 7 | Higher intelligence than Micro; still lightweight and fast. |
| Nova Pro | 8 | 8 | 8 | 8 | Good balance of intelligence, reasoning, and speed. |
| Claude 4.5 Haiku | 8 | 8 | 7 | 8 | Strong accuracy and reasoning; suitable for efficiency-focused use cases. |
| Claude 4.5 Sonnet | 7 | 9 | 9 | 9 | Flagship-level intelligence and reasoning. |
| Claude 3 Haiku (Deprecated) | 8 | 7 | 6 | 7 | Reliable and efficient for its time. Included for consistency, but other models are likely a better choice for modern applications. |
| Claude 3.5 Haiku (Deprecated) | 8 | 8 | 7 | 8 | Strong accuracy and reasoning; suitable for efficiency-focused use cases. |
| Claude 3.5 V2 Sonnet (Deprecated) | 7 | 9 | 9 | 9 | Flagship-level intelligence and reasoning. |
Complete each of these tasks in the order given.
Launch GenAI Prompts
-
Click the app selector
. -
Under General, select AI Studio.
-
In the AI Studio left navigation, select GenAI Prompts.
Create Prompt Profile
Required permissions: Launch GenAI Prompts, Prompt Access Edit
You will also need a license for NiCE CXone Copilot for Agents and the autosummary feature enabled.
Each flow comes with one default profile. You can create a custom prompt profile either by using a blank form or by copying an existing prompt and making edits. As you're working in the profile, it automatically saves for you only. Others cannot see the new profile, or edits you're making to an existing profile, until you click Publish Prompt. Profiles are also unavailable to select as the live prompt until they have been published.
If you create a new profile but don't publish it within 30 days of the most recent edit, the profile settings on the Prompt Configuration tab revert to their defaults. If you edit an existing profile but don't publish within 30 days of the most recent edit, changes on the Prompt Configuration tab revert to the settings specified in the latest published version.
- In the Profiles tree expand Copilot for Agents > Email AutoSummary.
-
Create a new, blank profile. To do so, click the plus icon
next to the Email Autosummary group. The new profile appears at the bottom of the list. -
Duplicate an existing profile.
-
Click the stacked dots icon
next to the existing profile you want to copy and select Duplicate. -
Rename the copy by locating it at the bottom of the AutoSummary profiles, clicking the stacked dots icon
, and selecting Rename. Enter the new name for the profile and click the check mark
to save it.
-
-
On the Prompt Configuration tab, adjust the values for these fields as needed:
-
Who is the contact center agent?
-
Who is the contact center customer?
-
What is the max length of the summary?
-
What is the language of the summary?
-
What are the instructions for the summary?
Learn more about fields in this step
Field Details Who is the contact center agent? A brief description of the agents handling the interaction.
For example, Customer service assistant for the business Classics, Inc..
This setting informs the generative AI prompt, but it doesn't directly influence the generated output. If you want to influence the output summary text, include written details in the Instructions field.
For example, When referencing the agent, use the word "Associate".
Who is the customer?
A brief description of the person being helped. For example, Classics, Inc. customer.
This setting informs the generative AI prompt, but it doesn't directly influence the generated output. If you want to influence the output summary text, include written details in the Instructions field.
For example, When referencing the customer, use the word "Consumer".
What is the max length for the summary? The maximum length of the summary between these options:
-
Short (1 - 50) words
-
Medium (51 - 100) words
-
Long (101 - 200) words
-
Extra Long (201 - 250) words
-
Extended (251 - 500) words
Large language models attempt to follow the selected maximum length, but the generated summary may be slightly longer or shorter than the requested range. This can be more noticeable for shorter lengths. For more information, see About GenAI Prompts response length behavior.
What is the language of the summary? The language the summary appears in. You can choose between these options:
- English (United States)
-
English (United Kingdom)
-
English (Australia)
-
English (Canada)
-
Portuguese (Brazil)
-
Portuguese (Portugal)
-
Spanish (Latin America)
-
Spanish (Spain)
-
French (Canada)
-
French (France)
-
German (Germany)
-
Dutch (Netherlands)
-
Japanese (Japan)
-
Italian (Italy)
-
Welsh (Wales)
-
Korean (South Korea)
-
Norwegian (Norway)
-
Finnish (Finland)
-
Chinese Simplified (China)
-
Chinese Traditional (China)
This field only impacts the language of the generated content.
Considerations for multilingual Automated Summary (AutoSummary) deployments
The large language model (LLM) selected for Automated Summary (AutoSummary) can affect summary generation in multilingual environments. When configuring Automated Summary (AutoSummary) for multiple languages:
-
Test representative interactions in each language your organization supports.
-
Verify that summaries are generated in the expected language.
-
Review generated summaries for accuracy and completeness.
-
Validate the configuration using the GenAI PromptsTest Suite before publishing changes.
Different LLMs can produce different results for multilingual interactions. When evaluating a model, test the languages and interaction types used in your organization to ensure that the generated summaries meet your business requirements. After publishing a prompt profile:
-
Allow time for the updated configuration to become available to users.
-
Monitor generated summaries to verify that the results meet your business requirements.
-
If necessary, update the prompt configuration and republish the profile.
For Automated Summary (AutoSummary) deployments that require summaries in languages other than English, Claude Sonnet is the recommended model. If your prompt instructions include custom redaction rules, Claude Haiku or Claude Sonnet are recommended. Test the selected model with representative multilingual interactions and prompt configurations before deploying changes to production.
What are the instructions for the summary The information you want the summary to include and the format you want it in.
Important: The What is the language of the summary? field only affects the language of the generated summary. You must enter these instructions in English.
For example:
###Instructions###
Summarize:
Summarize the intent, steps taken, and resolution of the interaction in a paragraph called Call Summary. If an order was placed, list the order details as a paragraph called Order Details. If no order was placed, exclude the Order Details paragraph. For Order Details, list the following details:
-Contact name
-Order recipient
-Order number
-Account number
-Contact phone number
-
-
Free text fields are used for short inputs such as:
-
Agent label
-
Customer label
-
Persona description
-
-
Text area fields are used for longer inputs such as:
-
Instructions
-
Custom prompts
-
-
About GenAI Prompts response length behavior
Large language models (LLMs) attempt to follow instructions to limit the length of their responses. However, the actual output can sometimes be slightly longer or shorter than requested. This behavior is expected and is not specific to a single model or feature. It occurs because:
-
Word counts: LLMs generate text based on tokens, not exact word counts. The mapping between tokens and words is approximate.
-
Natural language generation: The model prioritizes coherent, natural sentences. In some cases, completing a sentence or thought may exceed the requested length range.
-
Complex prompts or topics: When the interaction is complex, the model may use more words to include required details, even when asked to keep the response brief.
What to expect
-
The selected maximum length (for example, Short, Medium, Long) is treated as guidance, not a hard limit.
-
Minor variation around the selected range is normal.
-
Variance may be most noticeable when using shorter length options, because a few extra words make a larger percentage difference.
Recommendations
-
Use shorter length options when you want concise, high-level summaries.
-
If Agent Workspace (Agent) or Supervisor Workspace (Supervisor) require more detail, select a longer length option and adjust instructions to specify what must be included or excluded.
-
Periodically review generated summaries with your agents and supervisors and refine the prompt configuration (instructions, tone, and length) as needed.
-
GenAI Prompts Field Length Limits
GenAI Prompts fields support two types of text inputs. Each type has a defined maximum length.
|
field type |
Maximum length |
|---|---|
| Free text fields (for example, labels, persona descriptions) | 200 characters |
| Text area fields (for example, instructions, long prompts) | 10,000 characters |
The type of field used depends on the prompt configuration.
GenAI Prompts dynamically displays the appropriate field type based on the prompt configuration.
The user interface displays a warning only when the maximum length is exceeded. To avoid errors, ensure your inputs stay within the supported limits.
Publish the Prompt
Required permissions: Launch GenAI Prompts, Prompt Access Edit
After you have configured the prompt settings, click Publish Prompt. Publishing the prompt makes it available both for others to see and for selection in Copilot for Agents profiles.
Select the Prompt in Copilot for Agents
Once you have published the prompt profile, you can select it in the Copilot for Agents Config page profile.
- Click the app selector
and select General. - From the left menu, navigate to AI Studio > Profiles.
- From the list of Copilot profiles on the left, select the Copilot for Agents profile you want to apply the autosummary prompt profile to. If you don't have one yet, click Add Copilot Profile and follow these instructions to create one.
- On the Configuration tab, scroll down to the Email Channel section and turn on Generate Autosummary. From the Autosummary Profile drop-down, select the prompt profile you want to use.
- Click Save.