Re: [PR] Let Claude Agent SDK agent loops use common.ai toolsets through a new AirflowTools adapter [airflow]

["Lee-W (via GitHub)" <gi...@apache.org>]

Lee-W commented on code in PR #74026:
URL: https://github.com/apache/airflow/pull/74026#discussion_r4217027849


##########
providers/common/ai/docs/frameworks/index.rst:
##########
@@ -79,10 +79,17 @@ is Airflow's and which part stays yours.
        :doc:`../rag_pipelines`.
      - No agent
      - ``[llamaindex]`` extra
-
-The Strands and ADK integrations, the framework-neutral tool interface under them, and
-the tracing helper are experimental: they can change or be removed in a minor release of
-this provider. See :ref:`howto/stability`.
+   * - Claude Agent SDK, through ``HarnessOperator``

Review Comment:
   I changed into the same idea as #73962 and the Strands/ADK ones.
   
   The Dag author runs the Claude Agent SDK's `query()` loop, and `AirflowTools` serves common.ai toolsets to it through an in-process MCP server, with the same promises (masking, correctable errors to the model, everything else fails the task, metrics).
   
   ---
   
   The main difference from #73962 is that the SDK's loop is the Claude Code CLI, so a failure the model cannot fix fails the task one model request later than in yours. The PR description explains why



-- 
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: commits-unsubscribe@airflow.apache.org

For queries about this service, please contact Infrastructure at:
users@infra.apache.org