Repository navigation
fix: tell a chat once when a run's metrics are cut short - #1147
Merged
Merged
Conversation
Include the collection cause and saved-output guidance in the result turn when possible. Use each chat’s accepted prompt to avoid repeating the failure after collection settles or the app restarts. Show incomplete final collection on the run page alongside the last readings, without claiming the log contained no metrics. Assisted-by: gpt-6.2-sol Assisted-by: claude-opus-5.5
Contributor
[Medium impact] The PR appears safe to merge. SummaryThe PR avoids repeating a collection failure when a chat’s accepted result already included the same cause and saved-output guidance.
No actionable issues found. Tests and the app were not run during this review.
|
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
When a rental went away before its final log was kept, the chat's agent got two turns for it: the result notice already said why collection stopped, and then a second "signal collection could not finish" notice said it again. The follow-up now goes out only when the result the chat accepted didn't carry the cause (it was delivered before collection settled), including across a restart. The run page also says it couldn't finish reading the run's metrics and keeps the last readings, instead of "No metrics were found in this run's log."
Found running hand tests 60 and 61 live on orx compute rentals.
Assisted-by: gpt-6.2-sol
Assisted-by: claude-opus-5.5