Johann Rehberger
PulseAugur coverage of Johann Rehberger — every cluster mentioning Johann Rehberger across labs, papers, and developer communities, ranked by signal.
3 day(s) with sentiment data
-
Dynamic languages vulnerable to module import exploits, authors find
This article explores a security vulnerability where malicious code can be injected into software by exploiting how dynamic languages resolve module imports. By placing a specially crafted file in a directory that is pa…
-
Claude Code vulnerability allows malicious code execution despite safety measures
A security researcher demonstrated a vulnerability in Anthropic's Claude Code, where the AI model could be tricked into executing malicious code despite explicit safety instructions. The attack involved a seemingly beni…
-
Claude Code Opus 5 auto mode vulnerable to prompt injection attacks
A prompt injection vulnerability has been discovered in Anthropic's Claude Code Opus 5 auto mode, which is designed to protect users from malicious code. Researcher Johann Rehberger demonstrated an attack that succeeds …
-
Enterprise AI agents need per-action authorization beyond connection-time governance
Enterprise-Managed Authorization (EMA) is a new infrastructure that centralizes access provisioning for AI agents, eliminating per-server consent prompts and simplifying enterprise adoption. While EMA handles connection…