Research
An Exploration of Attacks on AI Agent Systems
Publication date unknown · Discovered arxiv.org
Page published
Publication date unknown · First observed: 14 Oct 2026
Coverage timeline
Single-source research — one report is available.
Why it matters
Prompt injection propagating across A2A-connected multi-agent systems shows how a single compromised agent can trigger cross-agent infections and data leakage, a growing risk as enterprises embed autonomous agents in real workflows.
A research paper from UCLA investigates how prompt injection attacks exploit both direct user inputs and inter-agent communication in single- and multi-agent AI systems, demonstrating system prompt leakage, bypassing restricted topics, unauthorized file system editing, and abuse of search capabilities. The authors extend these attacks to multi-agent frameworks using the A2A protocol and test an Agent Card Spoofing attack unique to that protocol, showing how weaknesses can propagate across agents.