LangGraph deployment via LangGraph Platform facing Permission error

[ashburncrash]

Hello team, I'm a LangGraph newbie :)

I've built a simple LangGraph agent and I've attempted to deploy it via LangGraph Platform. However, I'm facing below permission error and would like to find a way to resolve this:
google.api_core.exceptions.PermissionDenied: 403 Permission 'secretmanager.versions.access' denied for resource '<RESOURCE>' (or it may not exist).

I have my API keys stored in Google Cloud Secret Manager, and in my code base I retrieve them via google secrets API. Since I run gcloud auth application-default login, running the graph locally via langgraph dev has no issues.

My current method for retrieving secrets is following:

def get_secret(secret_name):
    name = f"projects/{GCP_PROJECT_ID}/secrets/{secret_name}/versions/1"
    response = client.access_secret_version(name=name)
    return response.payload.data.decode("UTF-8")

However, with deployment on LangGraph Platform, I am not sure how to configure a custom service account or grant access in a way. I'm open to hearing ideas 🙏 This will be important for building our product, so any help would be much appreciated.

Revision ID: 39fcc73d-d90c-46bd-8f28-54643942c75f

(I did search on discussion for similar issues but haven't found one. If there's an existing one then apologies!)

[ashburncrash]

I did a workaround of passing a service account's json key as environment variable when launching a deployment. But if there's a cleaner way, then feel free to share!