Security training gets blocked

“My issue is that the assistant often refuses or stops responding when I ask about standard, legitimate security topics, even on sanctioned wargames and labs.”

What gets counted gets built. Pile on.

Paste this into your agent. It files your request, with what you need it for.

I want this: rapsheet.dev/sheets/claude-code/issues/761/file
Follow it to file my request. Show me the text before you send.

No install. No account. Public and anonymous. The link asks your agent for one report, nothing else. What it sends

Growing. 1 report in the last 7 days and none the week before. Not fixed, and no vendor reply yet.

Stop blocking security learning and CTF exercises on training platforms. Reporters practicing CTFs, Hack the Box, OverTheWire Bandit, TryHackMe and PortSwigger labs see the assistant flag, refuse or stop responding.

Is this your request?

File my request

Where: security content filtering
Versions: 2.1.280 to 2.1.310
Platforms: Linux, macOS

Part of: Work past cyber safeguard blocks5 requests

Cyber flags on development prompts8 people
Turn off or exempt the safety classifier for approved work3 people
Maintain owned or client servers without cyber safeguard blocks2 people
Security testing on own projects2 people

Original reports3, newest first

Nov 3, 2026Reduce over-refusals for legitimate security education on sanctioned platforms#106920
Sep 29, 2026Allow CLI usage for authorized security training and CTF exercises#98237
Sep 25, 2026Reduce false positives in security content filtering for CTF discussions#97210