The new class of attack against coding assistants revealed in the period rests on a premise that usually goes unnoticed: it only pays off because these tools became part of development infrastructure.
Attackers choose targets by return. Investing in a technique that hijacks coding assistants presupposes many assistants running, with broad access, on machines belonging to people who hold production credentials.
That describes precisely what happened to the category in a short time. The tool came in through the door of individual productivity, without the review an infrastructure system would receive, and accumulated access in equal measure.
The figure confirming the scale appeared weeks later: one of those tools reached 5 million weekly users, a number describing routine incorporation rather than experiment.
The practical conclusion is about governance rather than tooling: if the assistant participates in the flow producing production code, it has to be treated as a production system, with inventory, scope and audit logging.