AI Security Threat Feed

MONAI: path traversal allows arbitrary file write

EverNoteLoader: XXE exposes host files in LangChain

llama-index: JSON parsing DoS via deep recursion

xgrammar: uncontrolled recursion in grammar parsing causes DoS

vLLM: RCE via eval() in Qwen3 Coder tool parser

picklescan: detection bypass allows malicious pickle exec

skops: joblib fallback enables RCE via model load

skops: RCE via MethodNode unsafe deserialization

skops: OperatorFuncNode type confusion → RCE

ExecuTorch: heap overflow in method load, RCE risk

llama_index: path traversal allows arbitrary file read

llama-index Papers Loader: XML expansion DoS

LlamaIndex Obsidian: symlink traversal exposes host files

jupyter_core: config hijack enables cross-user code exec

Label Studio: XSS enables unauthorized actions via CSRF

llama_index: DoS via uncapped recursion in web reader

LLaMA-Factory: RCE via torch.load() unsafe deserialization

picklescan: scanner bypass enables DNS data exfiltration

jupyterlab-git: command injection via malicious repo name

litellm: privilege escalation viewer→proxy admin via bad API key