add(core): safety check downgrade warning (#11964)

Add per-turn notice when a request is downgraded to a fallback model due
to cyber safety checks.

**Changes**

- codex-api: Emit a ServerModel event based on the openai-model response
header and/or response payload (SSE + WebSocket), including when the
model changes mid-stream.
- core: When the server-reported model differs from the requested model,
emit a single per-turn warning explaining the reroute to gpt-5.2 and
directing users to Trusted
    Access verification and the cyber safety explainer.
- app-server (v2): Surface these cyber model-routing warnings as
synthetic userMessage items with text prefixed by Warning: (and document
this behavior).
This commit is contained in:
Fouad Matin
2026-02-16 22:13:36 -08:00
committed by GitHub
parent 08f689843f
commit 02e9006547
12 changed files with 843 additions and 4 deletions

View File

@@ -63,6 +63,9 @@ impl Stream for AggregatedStream {
Poll::Ready(Some(Ok(ResponseEvent::ModelsEtag(etag)))) => {
return Poll::Ready(Some(Ok(ResponseEvent::ModelsEtag(etag))));
}
Poll::Ready(Some(Ok(ResponseEvent::ServerModel(model)))) => {
return Poll::Ready(Some(Ok(ResponseEvent::ServerModel(model))));
}
Poll::Ready(Some(Ok(ResponseEvent::Completed {
response_id,
token_usage,