SOURCE-LINKED INTELLIGENCE

When Malicious Instructions Persist: Persistent Memory Poisoning Attack on Harness-Based Agents

arXiv · AI, language, vision and robotics · article · Sep 12, 2026 · UTC

Harness design has transformed the development of LLM-based agents by integrating memory, tool use, and runtime control. However, this design also introduces security and privacy risks because malicious instructions from external sources may be written into persistent memory and persist across sessions. To study this risk, we propose PMPA, a Persistent Memory Poisoning Attack against harness-based agents. PMPA embeds malicious instructions into benign external sources and induces the victim agent to write them into persistent memory without directly accessing to the agent framework. Once store

Read original source ↗ Open in workspace

recordType

paper

region

Global