fix(agents-server-ui): hide empty reasoning blocks and interleave them with run items#4570
Merged
Merged
Conversation
…m with run items Two fixes for the reasoning stream UI: - Models that report reasoning without exposing the tokens (e.g. OpenAI codex models) open a reasoning row on thinking_start that never receives a delta. Filter contentless rows out client-side so no empty 'Thought' block renders. Encrypted (Anthropic redacted) rows are kept and still render their placeholder. - Reasoning blocks now interleave with the run's text / tool-call items by _timeline_order instead of all stacking above the whole response — in multi-step tool-using runs each block renders where the model emitted it (think, write, call tool, think, ...). ReasoningSection becomes ReasoningBlock (one block per row) with expand/collapse state lifted to AgentResponseLive so it survives remounts, same as before. Co-Authored-By: Claude Fable 5 <noreply@anthropic.com>
Contributor
Codecov Report❌ Patch coverage is
Additional details and impacted files@@ Coverage Diff @@
## kevin/reasoning-content #4570 +/- ##
============================================================
- Coverage 57.95% 34.03% -23.93%
============================================================
Files 369 224 -145
Lines 40622 19080 -21542
Branches 11506 6775 -4731
============================================================
- Hits 23541 6493 -17048
+ Misses 17006 12552 -4454
+ Partials 75 35 -40
Flags with carried forward coverage won't be shown. Click here to find out more. ☔ View full report in Codecov by Harness. 🚀 New features to boost your workflow:
|
Contributor
Electric Agents Mobile BuildLocal mobile checks ran for commit The EAS Android preview build was skipped because the |
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.
Summary
Two fixes for the reasoning-stream UI added in #4508 (note: that feature is on
kevin/reasoning-content, not yet onmain, so this PR targets the feature branch):No more empty thinking blocks. Some models report that they reasoned but never expose the tokens (e.g. OpenAI codex models) —
pi-adapter.tsdeliberately opens a reasoning row onthinking_starteven when no delta ever arrives, so the UI rendered a blank live block that settled into an empty▸ Thoughtrow.AgentResponseLivenow filters out rows with no content client-side. Anthropic redacted rows (encryptedset) are kept and still render their placeholder, and a genuinely-streaming block appears as soon as its first delta lands. Persistence is untouched — empty rows are still recorded (they can carry the encrypted payload that must round-trip to the model).Reasoning blocks interleave with the response instead of stacking at the top. Previously all of a run's reasoning rows rendered in one
<ReasoningSection>above every text/tool-call item, so in multi-step tool-using runs step-3 thinking appeared above step-1 output. Reasoning rows already carry the same_timeline_orderas text/tool-call rows, soAgentResponseLivenow merges both streams into one ordered render list — each block renders at the position the model emitted it (think → write → call tool → think → …). On an order tie (legacy rows without_timeline_order), reasoning sorts before output.Implementation
ReasoningSection→ReasoningBlock: the component now renders a single entry; expand/collapse state is lifted toAgentResponseLive(keyed by row key) so it still survives the block unmounting/remounting, same as before.ReasoningEntrygains anorderfield (sameTimelineOrderspace as run items).LiveRenderEntryunion +compareLiveRenderEntriescomparator; item-vs-item ties keep delegating tocompareLiveRunItems..rootwidth wrapper inReasoningSection.module.cssis gone — blocks are now direct children of theAgentResponseroot, which applies the same width treatment, so they align with text items.lastItemby identity instead of array index (the index no longer maps 1:1 once reasoning entries are interleaved).Test plan
pnpm typecheckclean inagents-server-uipnpm testinagents-server-ui(88 passed)🤖 Generated with Claude Code