[Bug] Anthropic API Error: Overly aggressive safeguards blocking legitimate security code review

Status Open
Reported on v2.1.235
Maintainer reply None cached
Activity 0 comments · opened Sep 2, 2026

Bug Description
False positive from cyber safeguards on a legitimate internal security code audit of my own repository (branch feature/audit-critical-fixes). Blocked twice: once for a teammate agent doing static security review, then on the main session running a shell command. No offensive tooling involved — only reviewing my own diff for input validation, secrets, and error handling.
Request ID: req_011CeeybdzV53BWSSiQdYUJY

Environment Info

  • Platform: darwin
  • Terminal: Apple_Terminal
  • Version: 2.1.235
  • Feedback ID: 884f44d5-8ec0-4d82-ba8f-1b94bb9f9700

Errors

[   API Error: Sonnet 5's safeguards flagged this message. Details: [cyber]
   Request ID: req_011CeeybdzV53BWSSiQdYUJY]

View original on GitHub ↗