[Feature Request] Reduce false positive rate in safety guardrails for benign code review requests
Status Open
Reported on v2.1.278
Maintainer reply None cached
Activity 0 comments · opened Sep 22, 2026
Bug Description
The safety guardrails are overly sensitive and trigger false positives on benign requests. For example, simply asking the model to check/review comments in code triggers the 'safeguards flagged this message' warning. Please adjust the safety thresholds so harmless coding tasks aren't unnecessarily blocked.
Environment Info
- Platform: linux
- Terminal: ghostty
- Version: 2.1.278
- Feedback ID: e4ecea80-ea6b-4b89-9f2f-6631557f3f98