Skip to end of metadata
Go to start of metadata

You are viewing an old version of this page. View the current version.

Compare with Current View Page History

« Previous Version 32 Next »

The OnLink AI Rovo Agent allows you to connect Rovo to any SaaS system. You can create what we call a Topic. A Topic is what you want to pull from the target system e.g. Job Details, Leave, etc. Topics can also be more generic e.g. Employee, Vendor, etc. Once a topic is defined, users can ask the agent for information by saying things like “get me job details for the assignee of this issue”.

Summary

Screenshot

Description

Add new topic

image-20241018-040837.png

Once you install OnLink AI, go to Apps - Manage your apps menu and click on OnLink AI in the left side menu. Then click on “Rovo Agent”. Here you can add new topics.

Templates

image-20241018-040655.png

Use the provided Templates to make it easy to configure your topics.

Security

image-20241018-041615.png

You can define who the authorized users are that can ask for information about other users. Only information asked about the requesting user is available by default.

Resolver

image-20241018-041828.png

If you want to enable user lookups i.e., @-mentions, assignee, reporter keywords, etc then you need to first use OnLink to load all employees to JSM Assets. You should, at minimum, load the reference ID (e.g. employee ID) of the employee in your HR system as well as the field users will ask for e.g. email or Jira User ID.

AQL for email or account Id look up can be set here. The “Return Attribute” is the name of the attribute that holds the employee reference ID.

HTTP

image-20241018-042133.png

Configure the HTTP endpoint by providing the method, URL and other details. You can also test the endpoint here.

Prompt Response

image-20241018-042218.png

When you Test the endpoint, the API Response is displayed. You can then configure what fields you want to return in the Prompt Response. Use curly braces to include the specific fields.

You can also use simple expressions (if/else) and calculations (dateDiff, diffNumbers, etc).

Here are a couple of examples:

"Message": "$dateDiff({HireDate}, {$today}) eq 0 ? 'Congrats' : ''"
"UpcomingLeave": "$dateDiff({LeaveStart}, {$today}) gt 10 ? 'No upcoming leave in next 10 days' : 'Leave planned in next 10 days'"

You can also include static text e.g. Suggestions that are displayed to user on what other things they can do with the agent.

Resolution Instructions

image-20241018-042839.png

Optional. Provide a link to a Confluence page that has instructions to resolve tickets. You can use the out of box Rovo agent e.g. Service Request Helper to either generate this text. This link is meant to be a curated set of instructions.


  • No labels