Fix red team scan output_path issue - prevent interim evaluation overwrites (#42024)

* Initial plan

* Fix red team scan output_path issue - prevent interim evaluation overwrites

Co-authored-by: slister1001 <[email protected]>

* Update CHANGELOG.md to document red team scan output_path bug fix

Co-authored-by: slister1001 <[email protected]>

---------

Co-authored-by: copilot-swe-agent[bot] <[email protected]>
Co-authored-by: slister1001 <[email protected]>

Author: Copilot

sdk/evaluation/azure-ai-evaluation/CHANGELOG.md

@@ -9,6 +9,7 @@
 
 ### Bugs Fixed
 
+- Fixed red team scan `output_path` issue where individual evaluation results were overwriting each other instead of being preserved as separate files. Individual evaluations now create unique files while the user's `output_path` is reserved for final aggregated results.
 - Significant improvements to TaskAdherence evaluator. New version has less variance, is much faster and consumes fewer tokens.
 - Significant improvements to Relevance evaluator. New version has more concrete rubrics and has less variance, is much faster and consumes fewer tokens.
 

sdk/evaluation/azure-ai-evaluation/azure/ai/evaluation/red_team/_red_team.py

@@ -2642,7 +2642,7 @@ async def _process_attack(
                     strategy=strategy,
                     _skip_evals=_skip_evals,
                     data_path=data_path,
-                    output_path=output_path,
+                    output_path=None,  # Fix: Do not pass output_path to individual evaluations
                 )
             except Exception as e:
                 log_error(self.logger, f"Error during evaluation for {strategy_name}/{risk_category.value}", e)