fix: gate openai reasoning downgrade on model switches (#1562) (thanks @roshanasingh4)

2026-01-24 07:58:04 +00:00
parent 3fff943ba1
commit c97bf23a4a
6 changed files with 239 additions and 41 deletions
--- a/docs/reference/transcript-hygiene.md
+++ b/docs/reference/transcript-hygiene.md
@@ -48,6 +48,7 @@ Implementation:

 **OpenAI / OpenAI Codex**
 - Image sanitization only.
+- On model switch into OpenAI Responses/Codex, drop orphaned reasoning signatures (standalone reasoning items without a following content block).
 - No tool call id sanitization.
 - No tool result pairing repair.
 - No turn validation or reordering.