A fetch Response body can be read only once, so clone it before consuming it twice

response.json(), response.text(), response.arrayBuffer() and response.blob() all drain the underlying stream and set bodyUsed to true. A second read throws a TypeError saying the body stream has already been read, which is the failure mode of an interceptor or logging wrapper that reads the body for a log line and then hands the same response to the caller.

response.clone() produces a second readable copy and must be called before any read starts. The cost is real: the browser buffers the body so both branches can be consumed at different speeds, and cloning a large download doubles memory use. Inside a wrapper, prefer reading once and reconstructing a new Response from the parsed value, or simply returning the parsed value, over cloning every response defensively.