Keep the final reply clean when checks continue or the verifier is skipped #126

Merged
weichen merged 1 commits from feature/final-reply-pollution-fix into main 2026-10-10 07:56:41 +00:00
Owner

Why: 实测 run_b520bf2e3037 及此后所有同源验证运行的最终回复被内部过程文本顶掉——事实核查修正轮的收尾文本与同源跳过提示成为用户可见答复,正式总结被折叠进过程时间线(2026-10-10 用户实测反馈问题 1)。

改动:

  • stop() 事实核查分支补 demoteLastText()(与其他分支对齐),续写理由要求模型重新给出完整最终答复;
  • 验证器不可用不再走 StopDecision.Notice 正文:同源跳过记通过态检查行、真失败记未通过态检查行,均留过程时间线;
  • 续写封顶提示同样改为未通过态检查行。

Evidence: go test ./... -count=1 全绿(32 包);新增两条回归测试(修正轮降级旧总结+可见文本只剩续写收尾;跳过提示为检查行且不泄漏正文);三个旧契约测试更新为断言新契约。真实模型前后对照见 Dev 任务卡。

Refs: Dev 任务 feature/final-reply-pollution-fix。

Why: 实测 run_b520bf2e3037 及此后所有同源验证运行的最终回复被内部过程文本顶掉——事实核查修正轮的收尾文本与同源跳过提示成为用户可见答复,正式总结被折叠进过程时间线(2026-10-10 用户实测反馈问题 1)。 改动: - stop() 事实核查分支补 demoteLastText()(与其他分支对齐),续写理由要求模型重新给出完整最终答复; - 验证器不可用不再走 StopDecision.Notice 正文:同源跳过记通过态检查行、真失败记未通过态检查行,均留过程时间线; - 续写封顶提示同样改为未通过态检查行。 Evidence: go test ./... -count=1 全绿(32 包);新增两条回归测试(修正轮降级旧总结+可见文本只剩续写收尾;跳过提示为检查行且不泄漏正文);三个旧契约测试更新为断言新契约。真实模型前后对照见 Dev 任务卡。 Refs: Dev 任务 feature/final-reply-pollution-fix。
weichen added 1 commit 2026-10-10 07:48:17 +00:00
Why: real run run_b520bf2e3037 showed the user-facing reply being displaced by
internal bookkeeping - the fact-check correction round's closing text plus the
same-source verifier skip notice became the visible answer while the actual
summary was folded into the process timeline (frontend splits at the last
process part). Same tail pollution reproduced on every same-model run since.
Evidence: go test ./... -count=1 green (32 pkgs); new regression tests cover
the demotion of the pre-check summary and the skip notice becoming a check
part; three old-contract tests updated to assert notices land as process
checks instead of final text.
Impact: stop() fact-check branch now demotes the standing summary (aligned
with the gate and cap branches) and asks the model for a fresh full reply;
verifier errors record as process checks (pass-state for same-source skip,
fail-state for real errors) instead of StopDecision.Notice text; the cap
notice likewise becomes a failed check line.
Refs: Dev task feature/final-reply-pollution-fix; diagnosis in
.task/feature-dual-speed-tasks.md lineage and session analysis 2026-10-10.
weichen merged commit c6912b63e7 into main 2026-10-10 07:56:41 +00:00
Sign in to join this conversation.
No Reviewers
No labels
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: weichen/Agent-System-Backend#126