Issuetrak 18.0 is here, and the star of the show here is AI integration, which we've paired up with the ability to create AI-generated Issue Summarization. This set of new features allows you to get a brief summary of any issue you have access to with the click of a button.
Issue Summarization works by sending the issue description, notes, timestamps, issue submitter, and note creator to the AI provider. The AI provider processes that information with the model selected, along with any prompt configuration passed to it, and then provides a brief summary in a text box at the top of the View Issue screen. For those of you that can't wait to learn more about this, skip ahead to the section below to learn more. And for those of you that aren't keen on having this feature operating in your instance of Issuetrak: Don't worry, it's disabled by default, and has no configuration built in. You have to purposefully enable AI and Issue Summarization and configure it before it is functional.
We've also taken strides in achieving parity between API v1 and API v2. We will very soon be able to say with confidence that API v2 can do everything API v1 can do, and more.
Keep reading to find out more!
AI Integration and Issue Summarization
Have you ever wanted a quick digest of a long-running issue? Or maybe an issue with tons of detailed notes in it? Our new AI Issue Summarization feature makes it a one-click affair to get a quick rundown of any issue you're currently viewing.
How It Works
When enabled, Agents in your site get a Summarize button at the top of any issue they're viewing. When they click on this, Issue Summarization gets to work by sending the issue description, notes, timestamps, issue submitter, and note creator to the AI provider. The AI provider processes that information using the model you selected during configuration, along with any prompts passed to it, and then provides a brief summary in a text box at the top of the View Issue screen.
Disabled by Default
Both the "AI" and the "Issue Summary" switches are disabled by default, and contain no provider configuration. In short: Issue Summarization cannot and does not work out of the box. Configuration is required in order for it to work.
Configuration
We've designed this feature from the outset to work with a wide variety of AI providers to suit your needs. The bulk of the of the configuration for AI integration is specific to the provider you selected. We'll go over what you need to enter in our new articles here and here, but here's a brief rundown:
Provider - Choose from cloud providers such as Google Gemini, AWS Bedrock, Azure Foundry, or Azure OpenAI. Want to use your own self-hosted provider, instead, such as Ollama? Choose OpenAI/Compatible and then enter the connection information.
Model - Once you've entered valid provider info and selected a model to use, you can enable Issue Summary and get to work with using it!
Prompt - We've pre-populated default prompt information that we think will result in a decent summarization for a broad range of issue content. But we recognize that it might not suit everyone's needs, so we've made it so that you can customize the prompt. Don't want a timeline of the issue? Take that part out. Want every issue summary to output in all caps? You can tell it to do so.
API v2 Endpoints
We've added a ton of functionality to API v2, getting us very close to parity with API v1's featureset:
Service Level Terms & SLAs
- Service Level Terms GET (all)
- Service Level Terms GET (individual)
- Service Level Terms POST (add)
- Service Level Terms POST (search)
- Service Level Terms PUT (edit)
- Service Level Terms DELETE (delete)
- Add Issue SLA information to Issue endpoints
Global Issues
- Global Issues Link to global
- Global Issues Global issue settings for issue
- Global Issues Promote to global
Issue Notes
- Notes GET (all issue notes for issue)
- Notes GET (individual note)
- Notes POST (create issue note)
- Notes POST (search - can exclude by note text)
- Notes PUT (update issue note)
Issue Attachments
- Attachments GET (All attachments for issue)
- Attachments GET (individual attachment)
- Attachments POST (Search)
- Attachments DELETE (Delete issue attachment)
Issue Tasks
- Update Issues Tasks Endpoint
- Issue Tasks Complete task
Core Issue Operations & Features
- Issues - Change status through PATCH (close/reopen)
- Issues - Close in APIv2 POST
- Issues - Add notification suppression for the Create Issue process (not the field on the issue)
Reference & System Entities
- Users - Change UserID
- Locations - GET (individual location)
- Departments - GET (individual department)
New Dataset: AI Token Usage
We thought that you might want to monitor usage of the new AI functionality, so we've made it possible to track usage of AI tokens with a new AI Token Usage dataset in Custom Reports. You can also view token usage by opening the Settings lightbox and clicking AI Token Usage in the top right.Deployment Tools
Deploying Issuetrak would fail with the error "Object reference not set to an instance of an object" if the default app pool had been deleted in IIS.
Identity Management
When attempting to save a User Property Mapping for an OIDC connection, clicking the checkmark would remove the line item, and a 400 error would appear in the browser's dev console.Incoming Email
When attempting to edit a mailbox, an otherwise-blank screen would display the message "Sequence contains no elements" when the corresponding "modified by" user for that mailbox had been deleted from the database.
Issue Hub
- In the new Issue Hub, a non-sysadmin who had "Can be assigned Tasks" permission, but not "Can view Tasks in Issues" could see the "As Task Assignee" view, but selecting it would fail with the message "Couldn't load results — Something went wrong while loading this view. Try refreshing, and contact your administrator if it continues." The view would never load for that user.
- It was not possible to filter a multi-select UDF in Issue Hub.
- It was not possible to filter by Location ID in Issue Hub.
Issue Management
The First Response time was being saved using the server's time zone instead of user time zone
License
Customers on a Teams license could not consistently disable the setting to disable periodic version checks on the license page.
Location
When there was no Location set for a user, and you tried to select the Location field on an issue before the Submitter, the Location would be blanked out when you selected the Submitter.
Logging
It was not possible to record both local application logging, and application logging transmitted via oTel. Issuetrak now supports having both in place.
Outgoing Email
When Outgoing Email was disabled, scheduled reports, queued CSV report exports, and escalation rule notifications were still being generated and queued. This would result in all of the queued emails being sent all at once if email was re-enabled later.
Password Policy
- The "hours" field for Self-Service Password Reset on the Password Policy screen would always revert to a '1' no matter what the value was set to.
- Password resets would fail for users when the following conditions were met:
- The setting "Require that new passwords do not match any of the last N passwords" is enabled.
- The account had a row in its password history containing a reference to an old/removed version of the password hasher designated as "HasherId = 2" in the database.
Scheduled Issues
Scheduled issues would not be generated as intended if there was a null value present for the "SubmittedById" for just one of the scheduled issues.
Special Characters
Curly quotes ” were not being normalized into ASCII quotes " when they were being stored in the database, resulting in problems with finding values that contained the normalized quotes.
User Interface
- Users without a personal Dashboard/Hub reload override setting couldn't log in when the global reload interval was a fractional value, such as 0.5.
- Certain combinations of browsers, screen resolutions, and zoom levels resulted in issue attachments not being visible when clicking on the paperclip icon on an issue.
