[Feature Request] Add prompt injection vulnerability protection guidelines for user input handling

Bug Description
As part of a security review of a personal project that assists with pull request reviews, I want to verify that the system cannot be compromised through prompt injection or comparable attack vectors. I believe my session was flagged during a security review.

Environment Info

• Platform: darwin
• Terminal: ghostty
• Version: 2.1.198
• Feedback ID: 5861f481-fbae-40ea-aa7c-102df4667628

Errors

This model's safeguards flagged this message. This sometimes happens with safe, normal conversations. Switched to Opus 4.8. Send feedback with /feedback or learn more