Feature Requests

Problem: When an AI intent is successfully completed (e.g., gathering information and sending a webhook to Rewst for an email block), we want the ticket to close automatically. Currently, even if we configure the AI's final reply to say "This ticket will now be closed," the system still appends the "Does this resolve your issue?" question. This prevents the ticket from closing and requires the user to reply or a technician to manually review and close it. This defeats the purpose of a zero-touch, automated workflow. Use Case Example: A user requests to block an email domain. The AI intent gathers the domain and sends a webhook to Rewst. The AI is configured to reply: "We've blocked [domain] as requested. This ticket will now be closed." Actual Behavior: The AI sends the message but also adds, "Does this resolve your issue?" The ticket remains open. Desired Behavior: The AI sends only the configured message, and the ticket status is immediately set to "Closed." Proposed Solution: Add a new option to the Intent configuration, such as a checkbox: [ ] Auto-close ticket on successful completion When this is enabled, the intent should run, send its final configured reply, and then immediately set the ticket status to "Closed" without asking for user confirmation.

As an MSP I would like to support my clients from their SMS.

Techs are frequently re-answering previously answered questions. Can CoPilot generate a recommended answer to a question based on previous answers?

Allow for rules to be set up where tickets meet a criteria, they can be excluded from further rules. eg. P1 security tickets are not retriaged when escalated. Scheduled tickets are not renamed or triaged. etc

Would like the AI agents to have memory persistence among a users prior requests. I could see how having it even across an entire org could cause problems if a customer has business units that are not privileged to the issues of others. But the agent seeing past ticket history and weighing that may be helpful to spot or address recurring problems, might even be nice for the LLM to flag recurring issues within a clients responses.

AI that can identify trends in client's ticket history. Many times "band-aid" fixes are applied to individual support sessions when those symptoms could be indicative of greater issues. Something that automates this process and potentially reports that to the account managers could go a long way in providing long-term resolutions.

When a client schedules a meeting through TimeZest, a note is added to the ticket containing "TimeZest scheduled it with" It would be great to have that trigger a change in status - I would use Scheduled for instance.

We are using Copilot to create intents, which generates a JSON file for import. However, Thread does not currently offer an option to import intents from a JSON file. Adding this feature would streamline the process and improve efficiency for teams leveraging AI tools to build service catalog intents.