fix(observability): defer logical LLM completion until validation

Signed-off-by: Alex Fournier <afournier@nvidia.com>
2026-07-19 09:00:27 -04:00
parent 55b7903cf7
commit 2c8fd38f2e
6 changed files with 165 additions and 5 deletions
@@ -1270,6 +1270,7 @@ def run_codex_stream(agent, api_kwargs: dict, client: Any = None, on_first_delta
                ),
                "retry_count": attempt,
            },
+            defer_logical_completion=True,
        )

        def _interrupt_or_superseded() -> bool: