fix(agents): drop orphaned thinking blocks when NO_REPLY text leaves an assistant message with only thinking content (#99620) (#99772)

* fix(agents): drop silent reasoning-only replay turns

Prevent NO_REPLY cleanup from leaving signed reasoning that merges into the next assistant tool-use turn and bricks strict-provider sessions.

Related: #99620

Co-authored-by: SunnyShu <shu.zongyu@xydigit.com>

* docs: refresh transcript hygiene map

Regenerate the docs map after documenting silent reasoning-only replay cleanup.

Co-authored-by: SunnyShu <shu.zongyu@xydigit.com>

---------

Co-authored-by: Peter Steinberger <steipete@gmail.com>
This commit is contained in:
SunnyShu
2026-07-21 15:34:15 +08:00
committed by GitHub
parent 5b3e01ba3e
commit 796b6fcdcf
4 changed files with 99 additions and 10 deletions
+1 -1
View File
@@ -9126,7 +9126,7 @@ Do not edit it by hand; run `pnpm docs:map:gen`.
- H2: Global rule: image sanitization
- H2: Global rule: malformed tool calls
- H2: Global rule: tool result pairing
- H2: Global rule: incomplete reasoning-only turns
- H2: Global rule: incomplete or silent reasoning-only turns
- H2: Global rule: inter-session input provenance
- H2: Provider matrix (current behavior)
- H2: Historical behavior (pre-2026.1.22)