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

• URL: http://arxiv.org/abs/2609.13889v1
• Date: Sat, 12 Sep 2026 11:27:53 GMT
• Title: When Malicious Instructions Persist: Persistent Memory Poisoning Attack on Harness-Based Agents
• Abstract summary: We propose PMPA, a Persistent Memory Poisoning Attack against harness-based agents.<n> PMPA embeds malicious instructions into benign external sources and induces the victim agent to write them into persistent memory.<n>Across all settings, PMPA achieves average Injection Success Rate (ISR) and Cross-session Attack Success Rate (C-ASR) of 73.7%/ 55.5% on OpenClaw and 66.9%/ 81.7% on Claude Code.
• Score: 17.320527773100732
• License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
• Abstract: 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 stored, the poisoned memory can be retrieved in later sessions, triggering additional malicious actions and causing privacy leakage. We evaluate PMPA on OpenClaw and Claude Code across different backbone LLMs, input modalities, and trigger scenarios. Across all settings, PMPA achieves average Injection Success Rate (ISR) and Cross-session Attack Success Rate (C-ASR) of 73.7%/ 55.5% on OpenClaw and 66.9%/ 81.7% on Claude Code, while preserving benign task performance on both systems. We further evaluate a targeted prompt-level defense and find that it can reduce memory injection in many settings, but provides limited protection once the persistent memory has been poisoned.

Related papers

• Memory Provenance Laundering in LLM Agents: A Non-Amplification Firewall for Persistent Memory [2.98985980237058]
  Provenance-Preserving Memory Fire wall (PPMF) is a lightweight memory that preserves platform-maintained provenance and authorizes tool calls by matching action risk to authority of action-relevant memories.<n>With intact platform-maintained provenance, confirmation, and risk labels, no evaluated unauthorized high-risk action passes the PPMF gate.
  arXiv  Detail & Related papers  (2026-07-31T08:46:27Z)
• When Claws Remember but Do Not Tell: Stealthy Memory Injection in Persistent Personal Agents [55.87577638514179]
  We study the threat as stealth memory injection, in which a remote black-box delivers a single email payload that must induce the agent to poisoned memory.<n>We propose MemGhost, a one-shot payload generation framework.<n>Across 56 held-out test cases, MemGhost achieves 87.5% end-to-end success on OpenClaw with GPT-5.4 and 71.4% on Claude Code SDK with Sonnet 4.6.
  arXiv  Detail & Related papers  (2026-07-06T15:08:58Z)
• MemVenom: Triggered Poisoning of Multimodal Memories in Web Agents [74.64265314956441]
  We propose a unified black-box attack framework that poisons graph-structured external memory with coordinated text-image evidence.<n>MemVenom achieves strong end-to-end attack success with minimal impact on benign performance, reaching up to 99.15% on GPT-5-family web agents.
  arXiv  Detail & Related papers  (2026-06-09T11:53:25Z)
• From Untrusted Input to Trusted Memory: A Systematic Study of Memory Poisoning Attacks in LLM Agents [3.3489120273076876]
  Memory is a core component of AI agents, enabling them to accumulate knowledge across interactions and improve performance.<n>We present a systematic study of memory poisoning in LLM-based agents.<n>We identify four memory write channels and nine structural vulnerabilities in model capabilities, system prompt design, and agent system architecture.
  arXiv  Detail & Related papers  (2026-06-03T01:04:13Z)
• Hijacking Agent Memory: Stealthy Trojan Attacks Through Conversational Interaction [3.809725488301918]
  Large language model (LLM) agents increasingly leverage long term memory to support persistent and autonomous task execution.<n>Existing memory poisoning attacks assume that injected content can be stored directly in memory, overlooking the selective extraction and rewriting stages.<n>Mechanistic analysis indicates that the attack exploits embedding-space anisotropy and shifts attention patterns.
  arXiv  Detail & Related papers  (2026-05-28T14:02:00Z)
• MemEvoBench: Benchmarking Memory MisEvolution in LLM Agents [78.95081012334116]
  Large Language Models (LLMs) with persistent memory enhances interaction continuity and personalization but introduces new safety risks.<n> contaminated or biased memory accumulation can trigger abnormal agent behaviors.<n>MemeEvoBench is the first benchmark evaluating memory safety in LLM agents against adversarial memory injection, noisy tool outputs, and biased feedback.
  arXiv  Detail & Related papers  (2026-04-17T07:29:52Z)
• Zombie Agents: Persistent Control of Self-Evolving LLM Agents via Self-Reinforcing Injections [57.64370755825839]
  Self-evolving agents update their internal state across sessions, often by writing and reusing long-term memory.<n>We study this risk and formalize a persistent attack we call a Zombie Agent.<n>We present a black-box attack framework that uses only indirect exposure through attacker-controlled web content.
  arXiv  Detail & Related papers  (2026-02-17T15:28:24Z)
• Memory Injection Attacks on LLM Agents via Query-Only Interaction [49.14715983268449]
  We propose a novel Memory INJection Attack, MINJA, without assuming that the attacker can directly modify the memory bank of the agent.<n>The attacker injects malicious records into the memory bank by only interacting with the agent via queries and output observations.<n> MINJA enables any user to influence agent memory, highlighting the risk.
  arXiv  Detail & Related papers  (2025-03-05T17:53:24Z)
• AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases [73.04652687616286]
  We propose AgentPoison, the first backdoor attack targeting generic and RAG-based LLM agents by poisoning their long-term memory or RAG knowledge base. Unlike conventional backdoor attacks, AgentPoison requires no additional model training or fine-tuning. On each agent, AgentPoison achieves an average attack success rate higher than 80% with minimal impact on benign performance.
  arXiv  Detail & Related papers  (2024-07-17T17:59:47Z)

This list is automatically generated from the titles and abstracts of the papers in this site.

This site does not guarantee the quality of this site (including all information) and is not responsible for any consequences.