fix: format AskUserQuestion tool_result per SDK spec using LLM answer…#54
Merged
Conversation
… parser
Use an LLM (haiku) to map the user's free-text IM reply to the structured
answers format required by the Agent SDK docs:
{"questions": [...], "answers": {"<question>": "<label>"}}
Previously the tool_result content was a plain string, which doesn't match
the documented response format. Falls back to the `response` field if the
LLM parse fails.
Also saves the `questions` array in `_pending_ask` so the parser has the
option labels as context, and removes a duplicate `import re`.
Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
hilr
approved these changes
Jun 16, 2026
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
2 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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
… parser
Use an LLM (haiku) to map the user's free-text IM reply to the structured answers format required by the Agent SDK docs:
{"questions": [...], "answers": {"": ""}}
Previously the tool_result content was a plain string, which doesn't match the documented response format. Falls back to the
responsefield if the LLM parse fails.Also saves the
questionsarray in_pending_askso the parser has the option labels as context, and removes a duplicateimport re.