[BUG] False positive - legitimate medical-education content flagged as unsafe

Preflight Checklist

• [x] I have searched existing issues and this hasn't been reported yet
• [x] This is a single bug report (please file separate reports for different bugs)
• [x] I am using the latest version of Claude Code

What's Wrong?

Summary: Claude Code repeatedly shows red safety warnings on routine medical-education
study material. This is a false positive.

Problem: When this content is processed, red safety/policy messages fire and interrupt
the work. None of it is operational or dual-use - it's textbook medical education content

Likely trigger: ordinary micro/virology vocabulary (pathogen virulence, toxins, immune
evasion, transmission, antiviral resistance) seems to pattern-match dual-use/bioweapon
categories, despite being routine medical education coursework.

Request: Please review and tune the classifier so legitimate medical education in this
domain isn't flagged. Happy to provide specific examples.

Environment: Claude Code on Windows; working with .pptx and .json study files.

What Should Happen?

Not an error

Error Messages/Logs

Request trigger safety guidelines.

Steps to Reproduce

Make material, especially large on the mentioned topics

Claude Model

Opus

Is this a regression?

Yes, this worked in a previous version

Last Working Version

_No response_

Claude Code Version

Claude Code 2.1.119

Platform

Anthropic API

Operating System

Windows

Terminal/Shell

PowerShell

Additional Information

_No response_