Merged
Conversation
josephjclark
approved these changes
Apr 21, 2026
Collaborator
josephjclark
left a comment
There was a problem hiding this comment.
Works great, thanks @hanna-paasivirta
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Short Description
Fixes streaming bugs introduced with the switch to structured outputs that led to job code/YAML not being streamed before the text answers.
Fixes #461
Implementation Details
EVENT:changeswasn't firing inworkflow_chatWhen we switched to structured outputs and dropped answer prefills, the buffered content became a fragment of an unclosed object like
{"yaml": "...",— which always fails to parse, sosend_changeswas silently skipped and yaml/code only arrived in the finalcompletepayload. Now, we strip the trailing comma + whitespace, append}to make it valid JSON, and extract the field by key.job_chatdoes this already (but this PR adjusts it slightly to clean up dead code and match the approach in the two services).The field-boundary delimiter was too literal
The code matched the exact string
'"text": "'/'"text_answer": "'with a hardcoded space. Anthropic's structured outputs emits JSON with inconsistent whitespace .workflow_chathappened to come back pretty-printed, butjob_chatcomes back compact ("text_answer":").job_chat's delimiter never matched, so text never streamed at all and everything appeared at once. We've now replaced the literal match with regexr'"text"\s*:\s*"'so whitespace doesn't matter.AI Usage
Please disclose how you've used AI in this work (it's cool, we just want to know!):
You can read more details in our Responsible AI Policy