MemoryDojo: Benchmarking Poisoning Through LLM Agent Memory

Reported by arXiv cs.CR (David H. Yang, Blake Bullwinkel, Amanda Minnich, Mohammad Mohammadi Amiri, Giorgio Severi)

Summary by Cyber Security News9 Nov 2026, 06:00 CET

About this summary. This is a short excerpt from the description that arXiv cs.CR published with its article. Cyber Security News did not report or verify the underlying story. Read the original: https://arxiv.org/abs/2611.09721

arXiv:2611.09721v1 Announce Type: new Abstract: LLM agents benefit from using memory systems to improve performance on long-horizon, multi-session tasks. However, memory introduces a critical attack surface. Malicious instructions encountered in one session may be stored and propagated through memory before being retrieved and acted upon in later sessions.

Read the full article at arXiv cs.CR →

Source attribution: headline and facts are from arXiv cs.CR (arxiv.org). Summary method: excerpt of the source description. See our source attribution policy.

Related coverage

MetaPI: Constructing Prompt Injection Benchmarks from LLM Agent Benchmark

An Agentic Multi-Agent Framework for System Log Anomaly Detection

Authorization Drift in LLM Agents: Evaluating Boundary Compliance in Multi-Step Tool Use

GuardrailAgent: An Agentic Framework for Jailbreak Defense

SECURITY AFFAIRS AI-CYBERSECURITY NEWSLETTER ROUND 7

Signed but Not Safe: Requirement Integrity in Multi-Agent LLM Handoffs