POISE: Position-Aware Undetectable Skill Injection on LLM Agents

Agent skills provide a lightweight mechanism for extending general-purpose agents, but their open format exposes them to skill-poisoning attacks. A practically dangerous injection must stay invisible: if executing the payload derails the user's legitimate task, the resulting failure signal invite...

The Hidden Threat in Plain Text: Attacking RAG Data Loaders

Large Language Models LLMs have transformed human-machine interaction since ChatGPT's 2022 debut, with Retrieval-Augmented Generation RAG emerging as a key framework that enhances LLM outputs by integrating external knowledge. However, RAG's reliance on ingesting external documents introduces new...

Inject Custom Code Into PE File: InfectPE

Inject Custom Code Into PE File Using this tool you can inject x-code/shellcode into PE file. InjectPE works only with 32-bit executable files. Why you need InjectPE? You can test your security products. Use in a phishing campaign. Learn how PE injection works. …and so on. In the project, there i...