Prompt-injection content planted in files ingested by AI agent tools

Ibrahim Saud@tektrix

•updated 13 days ago•1•0•2

Detects the creation or modification of documents in common formats (e.g., .txt, .md, .docx, .pdf) that contain strings indicative of prompt-injection attempts. These strings are commonly used to manipulate the behavior of AI agents or Large Language Models (LLMs) that may process or index these files automatically.