You have to edit the DM to see the values, but it's much easier to read it from the manual.
I'll presume you've read the CIM manual at https://docs.splunk.com/Documentation/CIM/5.3.1/User/Overview . What specific questions do you have about what you read (or couldn't find)?
CIM is not an Easy Button. That is, installing the app will not make your apps CIM-compliant. Instead, you must add aliases, calculated fields, and/or other KOs so your app will produce CIM-compliant data. The CIM manual lists the fields expected by each datamodel (not all fields are required).
Depending on your app, it's possible no DM will apply. That's OK.
Imagine I have an app with Intrusion Detection data
If I want to make my app CIM compliant I need to add aliases, tags and calculated fields like in the Intrusion Detection Datamodel ?
For example, if I have a field called "Alert level", I need to create an aliases in my app in order to rename it as "severity_id"?
Or is it better to create a own datamodel from my app and to query from this datamodel with tstats?
| tstats count from datamodel=TEST
Don't create your own DM. That defeats the purpose of CIM.
Your app should produce fields listed in the CIM manual for the Intrusion Detection model. It doesn't have to produce all of the fields, but as many as apply to the data. It also may have to adjust field values to match those expected by the DM ("high", "medium", "low", etc. in severity, for example). Tag the data as expected by the DM.
when you say "to match those expected by the DM ("high", "medium", "low", etc. in severity, for example)", where I can see this information in the DM?
You have to edit the DM to see the values, but it's much easier to read it from the manual.