Under the hoodTier: how much it should matter to you
1Useful: my rating, 1 to 5
2Signal: worth watching, 1 to 5
LLM GatewayArea: what it touches
Internal ChangesKind: in v2.1.295,
Internal ChangesSection of the release
What
In the gateway, the part that relays requests between Claude Code and the model service:
It reads a new x-gateway-upstream-request-id header and removes it from the request.
It sets a request-id header on successful replies.
It passes on a timeout for how long to wait for the first part of a reply, taken from the timeouts an operator has set.
Why
Requests through the gateway can be traced more easily, and its timeouts follow the operator's settings.
How sure we are
One source agreesOne thing we can check says the same as this entry.
Anthropic's release notes agreeAdded a request-id header to the Claude apps gateway's successful inference responses, so request_id in Claude Code telemetry matches the…
