Spark

Spark exposes conversation handoff through NexthinkClient.spark.

The handoff endpoint requires a target user principal name in the request headers. The client takes it as an explicit argument so scripts do not hide which user context is used.

Hand off a text message

from nexthink_api import NexthinkClient, NxtRegionName
from nexthink_api.Spark import (
    NxtSparkHandoffConversationMessageRequest,
    NxtSparkMessageDTO,
    NxtSparkTextPartDTO,
)

client = NexthinkClient(
    instance="example",
    region=NxtRegionName.eu,
    client_id="client-id",
    client_secret="client-secret",
)

request = NxtSparkHandoffConversationMessageRequest(
    message=NxtSparkMessageDTO(
        parts=[NxtSparkTextPartDTO(text="Summarize the current device health.")]
    ),
    metadata={"source": "automation"},
)

result = client.spark.handoff(
    request,
    user_principal_name="user@example.com",
    timezone="Europe/Paris",
)

user_principal_name is required. timezone is optional and is sent as the Spark Timezone header when provided.

Include file content

from nexthink_api.Spark import NxtSparkFilePartByContent

request = NxtSparkHandoffConversationMessageRequest(
    message=NxtSparkMessageDTO(
        parts=[
            NxtSparkTextPartDTO(text="Analyze this CSV."),
            NxtSparkFilePartByContent(
                fileContent="base64-encoded-content",
                mimeType="text/csv",
            ),
        ]
    )
)

See Spark object models for the request and response classes used by this domain.