[Bug] Cyber Verification Program status not honored in API requests, safeguard blocks legitimate learning plan requests

Bug Description
The Cyber Verification Program safeguard is blocking legitimate cybersecurity learning and career planning requests, despite the user being an active CVP participant. The safeguard should not apply to educational planning tasks.

Expected Behavior

• CVP participants should have safeguard exceptions applied to their account
• Legitimate requests about cybersecurity courses, learning paths, and career planning should not be blocked
• If a safeguard is triggered in error, the response should note CVP status and allow the request

Actual Behavior
Requests are blocked with a cyber safeguard error and instructions to "apply to the Cyber Verification Program," even though:

1. User is an active CVP participant
2. Requests are explicitly non-offensive (task management, learning planning)
3. User has clarified the use case multiple times

Steps to Reproduce

1. Log in as a CVP participant
2. Mention specific cybersecurity courses (e.g., "Offsec OSED EXP-301") in a learning/career planning context
3. Request modifications to a learning plan/timeline
4. Observe: safeguard error is returned despite CVP status

Environment Info

• Platform: linux
• Terminal: gnome-terminal
• Version: 2.1.258
• Feedback ID: 2a252246-7cfc-4756-b3d5-a89f1fef12e2
• Request ID: req_011Cef411WQFWA5cRB9hcEMu

Impact
CVP participants cannot use Claude for legitimate educational planning in their field of expertise, defeating the purpose of the verification program.