Skip to content

Commit 56db2d7

Browse files
Copilotslister1001
andauthored
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 <103153180+slister1001@users.noreply.github.com> * Update CHANGELOG.md to document red team scan output_path bug fix Co-authored-by: slister1001 <103153180+slister1001@users.noreply.github.com> --------- Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com> Co-authored-by: slister1001 <103153180+slister1001@users.noreply.github.com>
1 parent 0c44507 commit 56db2d7

File tree

2 files changed

+2
-1
lines changed

2 files changed

+2
-1
lines changed

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

Lines changed: 1 addition & 0 deletions
Original file line numberDiff line numberDiff line change
@@ -9,6 +9,7 @@
99

1010
### Bugs Fixed
1111

12+
- 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.
1213
- Significant improvements to TaskAdherence evaluator. New version has less variance, is much faster and consumes fewer tokens.
1314
- Significant improvements to Relevance evaluator. New version has more concrete rubrics and has less variance, is much faster and consumes fewer tokens.
1415

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

Lines changed: 1 addition & 1 deletion
Original file line numberDiff line numberDiff line change
@@ -2642,7 +2642,7 @@ async def _process_attack(
26422642
strategy=strategy,
26432643
_skip_evals=_skip_evals,
26442644
data_path=data_path,
2645-
output_path=output_path,
2645+
output_path=None, # Fix: Do not pass output_path to individual evaluations
26462646
)
26472647
except Exception as e:
26482648
log_error(self.logger, f"Error during evaluation for {strategy_name}/{risk_category.value}", e)

0 commit comments

Comments
 (0)