guan404ming opened a new pull request, #70070: URL: https://github.com/apache/airflow/pull/70070
## Why The human review panel could display something different from what the agent actually produced: JSON-looking text was shown as a Python dict and "true" became True, so reviewers approved content that didn't match the task's real output. ## How - Return XCom values as stored instead of re-parsing them in the HITL review plugin - Store plugin-written review data natively, matching how workers write XComs - Seed test XComs through the real worker write path so the mismatch is covered --- ##### Was generative AI tooling used to co-author this PR? - [X] Yes — Claude Code (Fable 5) -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
