No matter how you like to participate in developer communities, Pulumi wants to meet you there. If you want to meet other Pulumi users to share use-cases and best practices, contribute code or documentation, see us at an event, or just tell a story about something cool you did with Pulumi, you are part of our community.

Pulumi Community

you don't need the `get` property at the end there, I think

ya, tried without `get` which gives me of `Output` , this can't be passed to the `provider` property because `provider` is not `Input` type

You’re going to want your stack output to be the information you need to build the provider instead of the provider object

Sorry I don't fully understand your suggestions, I am using K8S Cluster Provider in the project A, therefore I want to reuse it in project B just like passing variables. Are you saying I must create a `provider` thing to achieve that?

Hi <@U01E602D5JA>
Thanks for pointing this out, well actually this is how I implement it in my first version.
If that's the only way to do so, then I must get information below to construct a `K8s.Provider` object, which totally makes sense as this is how I use it in project A anyway.
1. endpoint
2. masteAuth
3. cluster name
Since the `StackReference` provide access to output value, I was thinking of accessing the `ClusterProvider` object directly instead of constructing it the same way I did in project A.

Are you saying that it is not possible to directly reference an output from StackReference?

I’m saying look at the <https://github.com/pulumi/pulumi-kubernetes/blob/master/sdk/dotnet/Provider.cs|ProviderArgs class>. It accepts Input&lt;T&gt; values which means you can use Outputs to construct it.

So instead of outputting the Provider itself in Project A, output the _values that you need to construct it_ and then instantiate a new instance of the Provider in project B.

I see, that's exactly what I mentioned above. Although I know could definitely work for sure, doing so would have to create redundant object across different stack(which is not a big deal tho).

I was trying to directly use the output of project A in project B as if they are in a single project. But if you are indicating there is no way to do that, I am happy to use my first implementation then.

Thank you

The Provider object is not a resource that is represented in cloud so personally I see no problem with creating it twice. It’s more of an ephemeral object than anything

But in the event that it was a cloud represented object, I still would prefer the route of outputting an ID and doing one of the Get methods