Its theme's rank, 14 days: 2 on the Claude Code sheet
11 people reported
All time, since Oct 22, including 3 thumbs-up on GitHub.
Refusal on own-code hardening
“The model seems to treat “making RE harder” as a signal of malicious intent even when the context makes clear it’s the user’s own code.”
Is this your bug? What gets counted gets fixed. Pile on.
Paste this into your agent. It files your report, with your error and version.
I hit this bug: rapsheet.dev/sheets/claude-code/issues/23471/file Follow it to file my report. 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
Security refusal blocks protecting own code from reverse engineering. Requests to obfuscate or harden the user's own source code are stopped with a '[cyber]' refusal or security errors, despite stated intent.
Is this your bug?
File my report- Where
- Opus 5.5 model
- Versions
- 2.1.306 to 2.1.318
- Platforms
- Linux, macOS, Windows
- How bad
- Breaks a normal workflow.
Part of: Safeguards block session requests34 bugs
- Cyber safeguard blocks sessions377 people
- Reasoning_extraction safeguard block134 people
- Cyber safeguard pauses sessions52 people
- Safeguards refuse work requests48 people
- Safeguards block requests47 people
- Classifier halts routine replies12 people
- Requests blocked with a [cyber] flag8 people
- SSH to own devices refused7 people
- Pentest work blocked6 people
- CTF requests flagged as malicious6 people
Show all 33Show only the largest 10
- Classifier blocks own-project work6 people
- Classifier halts reverse engineering5 people
- Safeguards flag verified cyber work6 people
- Firmware requests flagged as cyber6 people
- Classifier stops benign replies5 people
- reasoning_extraction API error5 people
- Safeguard blocks all messages4 people
- [bio] classifier blocks responses4 people
- Reasoning extraction safeguard block4 people
- Security investigation requests blocked3 people
- Classifier stops DeFi bot work3 people
- Safeguards block CVP members3 people
- "[cyber]" flag stops sessions1 person
- Safety classifier blocks agent3 people
- Fictional hacking content flagged3 people
- Cyber verification prompt recurs2 people
- Cyber safeguards block own devices2 people
- Safety classifier stops auth work2 people
- Phishing research project flagged2 people
- Cyber stop on report tasks2 people
- Cyber safety block on audits2 people
- Biological safeguard triggers in sessions with no biology content2 people
- Unexpected safeguards notice2 people
Original reports8, newest first
- Request for anti-crack protection incorrectly flagged#108112
- Opus 5.5 Refuses VMProtect Analysis Requests and Auto-Downgrades to Opus 4.8#107373
- False-positive refusal on JavaScript obfuscation for source protection#107336
- Claude Code refuses to help obfuscate user's own code, citing IP protection concerns#107066
- False Positive Cyber Safeguard Blocks Software Licensing Work#107005
- Over-sensitive security filter blocks legitimate code hardening (anti-reverse-engineering) work#106765
- Safeguard false positive on production build hardening work#106306
- Opus 5.5 refuses code obfuscation requests on own source code with "[cyber]" stop#104374