7752e74e2b
update the code with review comments
2026-05-16 17:07:53 +08:00
ba99a23814
Merge branch 'main' into fix-2804
2026-05-16 09:27:40 +08:00
722c690f4f
fix(memory): isolate queued memory updates by agent ( #2941 )
...
* fix(memory): isolate queued memory updates by agent
* fix(memory): include user in queue identity
* Potential fix for pull request finding
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com >
* Fix the lint error
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com >
Co-authored-by: Copilot Autofix powered by AI <175728472+Copilot@users.noreply.github.com >
2026-05-15 10:26:35 +08:00
6ffe267d20
fix: use AIMessage content for summarization response parsing
...
Agent-Logs-Url: https://github.com/bytedance/deer-flow/sessions/0b34cccd-f69a-4f68-bfa6-9a41256b63b5
Co-authored-by: WillemJiang <219644+WillemJiang@users.noreply.github.com >
2026-05-11 14:48:45 +00:00
c995c3a394
fix: hide summarization LLM output from frontend during context compression
...
Fixes #2804
The useStream hook tracks messages-tuple mode which captures ALL LLM
events, including the internal summarization LLM call. This caused the
English summary text to briefly appear as an AI message in the UI before
the REMOVE_ALL_MESSAGES state update replaced it.
Fix by overriding _create_summary/_acreate_summary to pass callbacks=[]
when invoking the summary model, preventing LangGraph from forwarding
the internal LLM events to the frontend stream. Also add
hide_from_ui=True to the summary HumanMessage's additional_kwargs as a
belt-and-suspenders safety net alongside the existing name=summary
check.
2026-05-11 22:19:47 +08:00
881ff71252
fix(harness): preserve dynamic context across summarization ( #2823 )
2026-05-09 19:39:36 +08:00
5fd0e6ac89
fix(middleware): sync raw tool call metadata ( #2757 )
2026-05-08 10:08:53 +08:00
f9ff3a698d
fix(middleware): avoid rescuing non-skill tool outputs during summarization ( #2458 )
...
* fix(middelware): narrow skill rescue to skill-related tool outputs
* fix(summarization): address skill rescue review feedback
* fix: wire summarization skill rescue config
* fix: remove dead skill tool helper
* fix(lint): fix format
---------
Co-authored-by: Willem Jiang <willem.jiang@gmail.com >
2026-04-24 21:19:46 +08:00
4ba3167f48
feat: flush memory before summarization ( #2176 )
...
* feat: flush memory before summarization
* fix: keep agent-scoped memory on summarization flush
* fix: harden summarization hook plumbing
* fix: address summarization review feedback
* style: format memory middleware
2026-04-14 15:01:06 +08:00
Willem Jiang