Repository navigation
Keep narration out of answers; repair the reply tests the policy outran - #3804
Merged
Merged
Conversation
Narration before a tool call was stored as an AssistantTranscript, so it counted as an answer and could finish a record twice. Carries closed #3594's ad6a752: narration is NarrationTranscript, and each record finishes once. The 13 reds on main were stale tests, not lost replies: - Agent info messages get no automatic reply since 6964ac4; the commanded test now asks a question. - Only the turn a delivery opened answers it since 1973a5d; the reply helper now finishes that turn. - The bridge fixture declared viewer native; production bridges are external. - #monitor was never created. A running requester can't be deleted; it stops first. - StoredBinding returns the stored Binding, not its address. - The concurrency worker crashed on a duplicate factory argument before reaching the store. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XbFMSpSd8zQxyozgEV2Re7
There was a problem hiding this comment.
Copilot was unable to review this pull request because the user who requested the review has insufficient remaining quota. Try requesting a review at a lower effort level.
An organization owner can choose the repository organization as the billing source for reviews requested by members with a paid Copilot license. Learn about Copilot code review billing.
Claude Code usually saves reasoning under this tag with empty text. The conflict resolution dropped the old guard, so every such block finished an empty NarrationTranscript. It now takes the parent's hidden-thinking path. Caught by claude-board. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XbFMSpSd8zQxyozgEV2Re7
trissim
enabled auto-merge
October 31, 2026 06:14
|
debt-src at 4225075, every finding: |
Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Claude-Session: https://claude.ai/code/session_01XbFMSpSd8zQxyozgEV2Re7
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
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.
The reply and binding reds on main were stale tests; the traces found no lost reply. One real defect came along for the ride.
Source: carries closed #3594's ad6a752, which never landed. Narration before a tool call was an AssistantTranscript, so it counted as an answer and a record could finish twice. Now NarrationTranscript, finished once per record.
Why each test was stale is in the commit message, with the commit that changed the policy.
Proof: test_monitor_delivery, test_claude_automatic_reply and test_cli_error_family: 111 passed before merging main, 113 passed, 1 warning in 96.02s (0:01:36). No full suite. Not checked: a live bridged thread, and when the delete case went red.
🤖 Generated with Claude Code
https://claude.ai/code/session_01XbFMSpSd8zQxyozgEV2Re7