Added a new config: output.schema.infer.value= [false,true] An example use case is when publishing fulldocuments that have the same / similar shapes: publish.full.document.only=true output.json.formatter=com.mongodb.kafka.connect.source.json.formatter.SimplifiedJson output.format.value=schema output.schema.infer.value=true Note: A schema is inferred for each document outputted by the change stream. So each document could have its own schema definition. For Bson types that are not supported by Kafka eg: minValue, ObjectId a string value will be produced using the output json formatter.