[Bug] Model safeguards blocking legitimate security advisory composition
Status Fixed / completed
Reported on v2.1.220
Maintainer reply None cached
Activity 0 comments · opened Jul 29, 2026 · closed Jul 29, 2026
Bug Description
I get this issue, while composing a security advisory for decky-loader.
Opus 5's safeguards flagged this message. Our intentionally broad safeguards allow us to deliver more capabilities faster, but can sometimes flag legitimate coding, cybersecurity, and biology tasks. Switched to Opus 4.8.
Environment Info
- Platform: linux
- Terminal: WezTerm
- Version: 2.1.220
- Feedback ID: 08c5ce55-42e9-485b-abfc-45dd0e31e1f5
Errors
[]