GitHub user joeyutong added a comment to the discussion: 
[Discussion][Observability 2/2] Built-in Operational Metrics for Flink Agents

@weiqingy You're right—there is no Python normalization equivalent to Java 
`ToolResponse` today. I corrected my earlier reply and updated the proposal and 
PR documentation: Java maps an unsuccessful `ToolResponse` to failure, while 
Python maps resource preparation or invocation exceptions to failure and treats 
any normal return as success. We are retaining the Tool/MCP outcome metrics 
with these explicit runtime-specific semantics. Strict alignment is tracked in 
[#956](https://github.com/apache/flink-agents/issues/956) and will follow 
[#926](https://github.com/apache/flink-agents/pull/926). The Metrics 
implementation is now available in draft 
[#955](https://github.com/apache/flink-agents/pull/955).


GitHub link: 
https://github.com/apache/flink-agents/discussions/901#discussioncomment-17880640

----
This is an automatically sent email for [email protected].
To unsubscribe, please send an email to: [email protected]

Reply via email to