🐛 fix(agent-runtime): always persist assistant reasoning to DB#15687
Merged
Conversation
PR #13494 gated message reasoning persistence behind preserveThinking (agent chatConfig + model extendParams / qwen|zhipu fallback). That gate is only meant to control whether reasoning is replayed into the next LLM payload — applying it to the DB write dropped thinking content for every non-qwen/zhipu reasoning model in server-side agent mode: reasoning streamed live via stream_end but vanished after refresh. Restore unconditional reasoning persistence in messageModel.update and keep the preserveThinking gate only for state.messages payload replay. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
|
The latest updates on your projects. Learn more about Vercel for GitHub.
|
Member
Author
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 0779992e09
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Codecov Report✅ All modified and coverable lines are covered by tests.
Additional details and impacted files@@ Coverage Diff @@
## canary #15687 +/- ##
===========================================
- Coverage 80.93% 60.55% -20.39%
===========================================
Files 1536 3037 +1501
Lines 204910 265946 +61036
Branches 20622 30244 +9622
===========================================
- Hits 165847 161047 -4800
- Misses 38879 104885 +66006
+ Partials 184 14 -170
Flags with carried forward coverage won't be shown. Click here to find out more.
🚀 New features to boost your workflow:
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
💻 Change Type
🔗 Related Issue
Regression introduced by #13494.
🔀 Description of Change
#13494 added a
preserveThinkingfeature for Qwen3.7 Max, but it also gated the DB persistence of assistant reasoning behind the same condition (chatConfig.preserveThinking === trueand modelextendParamsincludespreserveThinking/ qwen-zhipu provider fallback).That gate conflates two distinct concerns:
state.messages). This should be gated by model capability, since most providers reject or ignore replayed thinking.messagestable for UI display. This should not be gated.Since the merge, every server-side agent run on a non-qwen/zhipu reasoning model silently lost its thinking content: reasoning streamed live via
stream_end, but after a page refresh it was gone (messages.reasoningwasnull). The interrupted-stream persistence path was never gated, making the happy path inconsistent with it as well.This PR:
reasoning: finalReasoningin themessageModel.updatecall.preserveThinkinggate only for thestate.messagespush (payload replay), renamingshouldPersistAssistantReasoning→shouldReplayAssistantReasoningto match its actual semantics.🧪 How to Test
Run a server-side agent with any non-qwen/zhipu thinking model (e.g.
lobehub/deepseek-v4-pro), let it produce reasoning, then refresh the page — the thinking block should still render (previously it disappeared).bunx vitest run 'apps/server/src/modules/AgentRuntime/__tests__/RuntimeExecutors.test.ts'All 116 tests pass.
📝 Additional Information
Behavior matrix after this fix:
reasoning🤖 Generated with Claude Code