Support teams use custom fields to organize and track requests like categorizing them by category, priority, or product area. But filling these fields usually means reading through each message and updating them manually. It’s repetitive, easy to miss, and adds overhead to your support. 

Last year we introduced AI generated Category and Sentiment fields in ClearFeed that were automatically populated. Today, we’re excited to introduce a more comprehensive framework for AI Fields in ClearFeed - that gives users complete control over the AI prompts. You can define your own prompts for extracting information from a Support ticket and populating custom fields like Product Area, Urgency & Category.

What this means for you:

• Reduce manual work by auto-filling fields based on the message
• Ensure consistent tagging across requests
• Surface key details early so teams can respond faster

Let’s look at how AI Fields work and how you can use them to streamline request management in Slack.

‍

How AI Fields Work 

AI Fields in ClearFeed let you automate the process of tagging or categorizing requests efficiently by using a language model to automatically fill in a field based on the content of a request. You write a prompt once, and ClearFeed runs it every time the automation is triggered.

Here’s how to set it up:

1. Create a Custom Field

Start by defining a field (e.g. “Request Type”) and its possible values (like Bug, Feature Request, Billing). You can choose from supported types like Text, Single Select, or Multi Select and enable the “Auto-Fill with AI” option on the field

‍

2. Customize the Prompt

As soon as you enable the ‘Auto-Fill with AI’option, a default prompt will be generated that tells the AI how to decide the value for this field. You can edit the prompt according to your needs and reference parts of the conversation using variables like {{request.all_messages}} or {{request.title}}.

‍

3. Preview & Test

You can test the prompt using real sample requests. During testing, ClearFeed shows you the AI’s selected value and its reasoning allowing you to refine the prompt before going live. 

‍

4. Set Up Automation

Use ClearFeed Automations to decide when the AI should run. For example: "When a request is created, auto-fill the field with AI ‘Issue Category". ClearFeed runs the prompt every time the automation is triggered, uses the prompt and populates the field based on the output. 

‍

Key Use Cases for Custom AI Fields 

1. Categorize Request Type

Automatically tag incoming requests as a Bug, Feature Request, or Billing Issue, based on the language used by the customer. This helps triage teams route issues correctly without reading every message manually.

Example:
“Categorize this request as one of: Bug, Feature Request, or Billing. Use the conversation history to decide.” 

‍

2. Identify Urgency in Requests 

When a request is assigned to an agent, ClearFeed can analyze the message content and update an Urgency field. This makes it easier to spot time-sensitive issues without waiting for manual tagging.

‍

3. Generate a Request Summary 

You can use AI to fill a text field with a brief summary of what the request is about. This helps when reviewing new requests or syncing with tools like JIRA, where a quick description is useful.

‍

4. Auto-Categorize by Product Area

If your team supports multiple product areas like Billing, Reporting, or Integrations, it can be useful to tag incoming requests accordingly. With AI Custom Fields, you can automate this categorization based on how the customer describes the issue.

It helps route the request to the right team and makes it easier to filter and report on issues by product area.

‍

5. Detect Sentiment or Tone

Teams can define a custom field like “Sentiment” and let AI populate it with values like Neutral, Frustrated, or Positive. This adds an extra layer of context to each request and helps support teams proactively handle sensitive conversations.

‍

6. Flag Escalation Risk Early

You can create a custom field like ‘Escalation Risk’ and use AI to auto-populate it based on language that signals urgency, repeated follow-ups, or dissatisfaction. This helps support teams flag high-risk requests and bring in L2/L3 support managers before the issue escalates.

‍

Chaining Automations to Trigger Actions

Many use cases involve triggering actions based on AI-powered categorization. For example:

• Detect if a new Ticket is marked as ‘Critical’
• Notify the Customer Success Manager if it is

Users can build such workflows by chaining automations. In this case:

• Define an AI-field named ‘Urgency’ and populate it during Ticket creation
• When the Urgency field is updated, trigger another automation that posts a Slack DM to the CSM if the field value is Critical. 

Sophisticated automations like this can be built using the ClearFeed Automations Framework, which supports a range of triggers, conditions and actions.

‍

Wrapping Up 

AI Custom Fields are designed to reduce repetitive work and bring structure to unstructured support conversations without relying on manual tagging, saving time and ensuring accuracy. 

At this time this feature is in limited preview. If you’d like to learn more about ClearFeed's AI Fields, reach out to us at support@clearfeed.ai or book a free demo with a support specialist for a personalized walkthrough.