Skill UI

A comprehensive expert playbook detailing advanced Large Language Model (LLM) prompt injection techniques. Covers direct methods (instruction override, role play), indirect attacks (RAG poisoning, web data manipulation), and sophisticated tool/function calling abuse for data exfiltration and security bypass. Essential for rigorous AI application security testing.

This skill provides a comprehensive guide for authors targeting the International Semantic Web Conference (ISWC). It details the best practices for framing semantic web, knowledge graph, and ontology papers. It helps authors ensure their manuscript meets the venue's high standards regarding semantic interoperability, real-world data application, and specific contribution types, optimizing the submission strategy.