[Feature Request] Allow Claude Fable-5 to assist with security vulnerability remediation in user-owned applications

Bug Description
Fable 5 [1m] blocks legitimate security hardening in own application.
When using Fable 5 [1m] to address security vulnerabilities in my own greenfield project — specifically prompt injection and similar issues — the model switches to claude-opus-4-8 citing: "Fable 5's safety measures flagged this message for cybersecurity or biology topics."
This behavior seems overly broad. The trigger appears to be topic keywords rather than actual intent or context. Helping a developer secure their own application is a fundamentally different use case from assisting with attacks on external systems.

Expected behavior: Fable 5 should be able to assist with defensive security work within a user's own codebase without falling back to a different model.

Actual behavior: Any mention of prompt injection or related vulnerabilities triggers a model switch, regardless of context.

⏺ Fable 5's safety measures flagged this message for cybersecurity or biology topics. They may flag safe, normal content as well. These measures let us bring you Mythos-level capability in other areas 
  sooner, and we're working to refine them. Switched to Opus 4.8. Send feedback with /feedback or learn more: https://support.claude.com/en/articles/15363606
  ⎿  Tip: You can configure model switch behavior in /config

Environment Info

• Platform: darwin
• Terminal: Apple_Terminal
• Version: 2.1.170
• Feedback ID: 2713c981-7f8e-4298-b90a-2118643d4b68

Errors

[]