Plugin information
| Version | 0.0.9 |
|---|---|
| Author | Dataiku (Alex Bourret) |
| Released | 2025-04 |
| Last updated | 2025-05 |
| License | Apache Software License |
| Source code | Github |
| Reporting issues | Github |
This Dataiku DSS plugin provides a read connector to interact with ServiceNow tables, and an agent tool to allow LLMs to create ServiceNow issues.
How to set up
Create service accounts
As a DSS admin, go to Applications > Plugins > Installed > ServiceNow > Settings > Service accounts > + Add preset. Add the URL to the ServiceNow instance to reach. Add the user name and password of the ServiceNow account that will be used to interact with the instance. Finally, share the preset with the relevant groups.
Create an instance preset for user accounts
As a DSS admin, go to Applications > Plugins > Installed > ServiceNow > Settings > User accounts > + Add preset. Add the URL to the ServiceNow instance to reach. Share the preset with everyone.
Once this preset is create, each DSS user that requires accessing to the ServiceNow instance will have to go to their own profile page > Credentials > created preset > Edit symbol, and add their own user name and password.
How to use
Read a ServiceNow table
In the flow, click on +Dataset > Plugins/ServiceNow > ServiceNow table to dataset. Select the appropriate authentication type and preset and the table to retrieve. Press Test & get schema and Create
Use with a LLM agent
This plugin can be used as a tool by an agent to create ServiceNow issues. This requires Dataiku >= 12.6.2.
Example Use Case: Create Issues Based on User Input
In this example, we aim to process a dataset of messages from user tickets and create ServiceNow issues if necessary.
- First, we need to create the agent tool. In the project’s flow, click on Analysis > Agent tools > +New agent tool > Create a ServiceNow issue. Name the identifier and the agent.
- Then, we can create the actual agent. Click on +Other > Generative AI > Visual Agent, name the agent, click on v1, and add the tool (+Add tool) created in the previous step. In the agent’s Additional prompt section, describe what are your specifications on what the parameters in the ticket should contain. The agent tool can fill in the following parameters: description, summary, emergency and impact. An example of specification could be The impact should be set to 1 only if the problem impact production
- Once this is done, the agent can be used in place of an LLM in your prompt and LLM settings. For instance, a simple LLM recipe with the prompt You are an IT agent and your mission is to write ServiceNow ticket if the user’s message requires it.