Closed
Conversation
(cherry picked from commit a9f96653082321901f2967474c8d9b5cfec8b6ee) Update package.json Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Update src/middleware/validation.ts Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> Update package.json Co-authored-by: Copilot <175728472+Copilot@users.noreply.github.com> fix: adress review comments
…jects Move the client.chat.completions.create() call inside the existing try/catch/finally block so that if it rejects (e.g. auth error, network timeout, invalid model), the span is properly ended and the error is recorded. Previously the span would leak because the finally block never ran. Add regression test verifying recordException, setStatus, and end() are called on the span when create() rejects.
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
3 participants
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Currently, the inference loop will iterate over
handleOneTurnStreamand finish if one of either happens: (1) the MAX value (hardcoded to 5) happens; or (2) the payload.messages.length does not grow, which means there were no MCP calls needed to be resolved and added to the payload.That causes a problem in two cases, however:
responses.jsthen will pick that up, resolve the MCP call and rerun an inference, causing partial answersThis PR fixes it by adding a new loop breaker:
hasUserTask, which signals if there are any function/mcp calls that are demanding an user integration.If that is true, the loop must break and return the partial answer to the user.