Freshdesk Freshdesk

Resources

Products

[Early Access] Get solution article suggestions with Copilot Agent

Modified on: Thu, 22 Oct, 2026 at 6:04 PM

TABLE OF CONTENTS

Solution article suggester, a built-in Copilot Agent feature, recommends relevant knowledge base articles when an agent opens a ticket, so agents can find and share proven answers without searching the knowledge base manually. This article explains how solution article suggester recommends articles in Freshdesk, how to scope suggestions by folder visibility and article tags, and how agents can review and use the suggestions.


Prerequisites

Before you configure or use solution article suggester:

  • Ensure your account has the Freddy AI Copilot add-on on a supported plan.
  • Ensure you have administrator access to AI Agent Studio to configure solution article suggester.
  • Ensure your knowledge base contains at least 25 published solution articles.
  • Ensure agents have a Freddy AI Copilot license assigned.

Overview of solution article suggester

Solution article suggester reads the ticket subject and description and surfaces the articles most likely to resolve the issue. By default, Freddy AI considers all articles in the portal mapped to the ticket.


You can narrow the articles Freddy AI draws from. This is useful when one portal contains articles for multiple brands, product lines, or agent groups, and not every article is relevant to every ticket.


Solution article suggester provides the following capabilities:

  • Contextual suggestions: Recommends articles based on the ticket subject and description, with a visibility label and a confidence score for each article.
  • Folder visibility scoping: Limits suggestions to articles in folders with the visibility types you select.
  • Article tag mapping: Filters suggestions by matching the value of one ticket field to solution article tags.
  • Reply integration: Lets agents insert an article or its link into a reply.

Solution article suggestions work on newly created tickets from Email, Portal, Web Form, Web Chat, Phone, WhatsApp, Instagram (DMs and comments), Facebook (DMs and comments), SMS, and custom sources, including tickets created through the API.


Enable solution article suggester

To enable solution article suggester:

  1. Go to AI Agent Studio > AI Copilot.
  2. Under Smart Suggestions & Response Assistance, toggle ON Solution article suggester.

Configure the article suggestion scope

To configure solution article suggestion,

  1. Go to AI Agent Studio > AI Copilot.
  2. Select Customize next to Solution article suggester.
  3. In the left pane, under Build, select Configuration.
  4. Under Refine suggestions by folder visibility, select the folder visibility types Freddy AI can use.
  5. Under Map article tags for article suggestion, select the ticket field to match with article tags.

  6. Save and publish your changes.


You can use folder visibility, tag mapping, or both. If you configure neither, Freddy AI considers all articles in the portal mapped to the ticket.


Refine suggestions by folder visibility

Under Refine suggestions by folder visibility, choose the folder visibility types Freddy AI includes when it suggests articles.


Visibility type

Articles included

All Users

Articles in folders visible to all users, including customers and agents. Always enabled.

Bots

Articles in folders accessible to bots. Always enabled.

All Agents

Articles in folders restricted to agents only.

Only logged-in users

Articles in folders restricted to authenticated users.

Company, Contact & Contact segment

Articles in folders with contact or company segment visibility.


Toggle each visibility type ON or OFF to include or exclude the corresponding articles.


Map article tags to a ticket field

Tag mapping connects one ticket field to solution article tags. When a ticket arrives, Freddy AI reads the value of the selected ticket field and suggests only articles tagged with a matching value.


Tag mapping needs two things in place before you configure it:

  • Article tags: Tags on the solution articles you want Freddy AI to suggest.
  • An automation rule: A ticket creation rule that sets the selected ticket field to a value that matches those tags.


To create the automation rule,

  1. Go to Admin > Workflows > Automations > Ticket creation.
  2. Select New rule.
  3. Under Conditions, define what identifies the ticket. For example, Email address contains @acme.com, Subject contains Billing, or Product is Freshdesk. Combine conditions with Match ALL or Match ANY.
  4. Under Actions, select Add action, then select Set <ticket field name>.
  5. Choose the ticket field you plan to map, such as Product or a custom field like Delivery.
  6. Set the field value to the tag on the relevant articles, such as Standard delivery or Express delivery.
  7. Select Save.


To map the ticket field,

  1. On the Configuration page, go to Map article tags for article suggestion.
  2. Under Choose ticket field, select the field your automation rule sets. You can select one field: a default field, such as Group, Product, or Status, or a custom ticket field.
  3. (Optional) Select Consider articles without tags within selected folders or portal while finding relevant suggestions to include untagged articles alongside tagged ones.
  4. Save your changes.


The tag value must exactly match the tag on the articles. If the ticket field is empty, or no article tag matches, Freddy AI matches the ticket title and description to articles instead.

Default ticket fields apply from the same day. Custom ticket fields can take 4 to 5 days to become active for filtering.


Use solution article suggestions on a ticket

When an agent opens a ticket, Freddy AI analyzes the ticket content and surfaces relevant solution articles in the Freddy Suggests card below the ticket description. The card shows the number of solution article suggestions alongside similar ticket and canned response suggestions.


To use a suggested article,

  1. Open the ticket on the ticket details page.
  2. In the Freddy Suggests card, select Solution articles.
  3. Review the articles listed under Freddy Recommendations. Each article shows its title, visibility, and confidence score.

  4. To preview an article, select View article.
  5. Hover over an article and choose how to use it.

Action

What it does

Insert article

Inserts the full article text into the reply editor, below any content you have already typed.

Insert link

Inserts a hyperlink to the article in your reply.

Open in new tab

Opens the article in a new browser tab.


Find an article that is not suggested

If the suggestions do not cover the issue, agents can search the whole knowledge base from the same panel:

  • Search the panel: Use the Look for a solution article search bar at the top of the panel.
  • Use the reply editor: Type /s and press Enter, or select Suggested solutions at the bottom of the reply section.

Things to keep in mind

  • Supported tickets: Suggestions work on newly created tickets from all supported channels, including custom sources.
  • Knowledge base size: Knowledge bases with fewer than 25 articles may produce fewer or less accurate suggestions.
  • Agent access: Freddy AI suggests private or restricted articles only to agents who can access them, so two agents on the same ticket may see different suggestions.
  • Automation timing: Tag mapping applies only when the ticket field is set at the time the suggestion runs. Otherwise, Freddy AI uses subject and description matching.
  • Custom field activation: Custom ticket fields selected for tag mapping can take 4 to 5 days to become active.
  • Untagged articles: Including untagged articles broadens the results and may add less relevant suggestions.


The availability of optional functionality is subject to feature-specific terms and conditions in the Freshworks Supplemental Terms.