Tag
This repository provides artifacts for Segmentation Fault-Oriented Programming (SFOP), a novel exploitation technique that abuses signal handlers to bypass Intel CET. It includes PoC exploits and demonstrations against Nginx and Ladybird.
The author built a daily digest POC using IBM Bob, IBM's AI coding agent, and shares their experience with its new v2 features including parallel tool calls and simplified modes, while noting limitations like lack of reusable skills. They ask the community for their own experiences with Bob.
QLLM introduces a novel architecture without transformer or mamba, achieving O(1) inference and no KV cache. A 100M parameter model is released as a proof of concept, with open-source code.
A security researcher discovered severe vulnerabilities in MSI Center that allow any authenticated user to gain SYSTEM privileges via a named pipe service, enabling registry modification, WMI control, and arbitrary code execution as LocalSystem.
Building a PoC called 'Agent Harness Ops' using DeepAgents by LangChain, with observability via Langfuse, task queue with Celery/Redis, PostgreSQL, React frontend, deployed on Railway. Plans to share and write about it.
This paper details a novel bootROM vulnerability in Apple A12/A13 SoCs, exploiting a hardware bug in the USB controller and a configuration flaw to achieve boot-chain compromise. A proof-of-concept is provided.
The author proposes Agentra, a DNS-like system for AI agents to register, find, and invoke each other via API, simplifying multi-agent integration.
A new Linux privilege escalation exploit called Fragnesia has been released with proof-of-concept code on GitHub.