Red Agent: AI-Powered Command Execution
Red Agent is an LLM-driven component of RedC2 that translates natural-language prompts into framework beacon commands, enabling operators to execute complex post-exploitation…
Tag: LLM
Perplexity Deep Links Used in AI Recommendation Poisoning
Perplexity is among the AI assistants targeted by 'Ask AI' buttons that carry hidden memory-poisoning prompts, as observed in a consent management…
AI Recommendation Poisoning: How ‘Ask AI’ Buttons Silently Alter LLM Memory
A new class of prompt injection attack, dubbed "AI Recommendation Poisoning," is spreading across commercial websites. It exploits pre-filled deep links in…
ChocoShell (CHERRYPIE): PowerShell Infostealer with LLM-Generated Code
ChocoShell, also known as CHERRYPIE, is a PowerShell-based infostealer delivered via ClickFix lures. It steals browser session cookies, saved passwords, Microsoft 365…
GGUF: LLM Model Format
GGUF, the current standard for locally deployed LLMs, is among the AI file types targeted by ENCFORGE ransomware.
GGML: Legacy LLM Model Format
GGML, the predecessor to GGUF, is among the AI file types targeted by ENCFORGE ransomware.
Exposed Server Reveals AI-Assisted Phishing Toolkit Behind WebDAV Malware Campaign
A malware operator left its delivery server exposed, allowing Rapid7 to recover a full toolkit of 1,048 files including lure templates, filename-spoofing…
GPT-5.2
GPT-5.2 is a large language model by OpenAI. It was tested and found vulnerable to Agent Data Injection (ADI) attacks.
GPT-5-mini
GPT-5-mini is a smaller language model by OpenAI. It was tested and found vulnerable to Agent Data Injection (ADI) attacks.
Sonnet 4.5
Sonnet 4.5 is a language model by Anthropic. It was tested and found vulnerable to Agent Data Injection (ADI) attacks.