mlsanityData to 14 Oct 2026

Logic layer Prompt Control Injection (LPCI): A Novel Security Vulnerability Class in Agentic Systems

arXiv 2507.10457 ↗First posted 14 Jul 2025cs.CR cs.AI cs.LGcited by 6

On the map: LLM agents and tool use → Agent threats, guardrails and governance

Abstract (arXiv)

The integration of large language models (LLMs) into enterprise systems has introduced a new class of covert security vulnerabilities, particularly within logic execution layers and persistent memory contexts. This paper introduces Logic-layer Prompt Control Injection (LPCI), a novel category of attacks that embeds encoded, delayed, and conditionally triggered payloads within memory, vector stores, or tool outputs. These payloads can bypass conventional input filters and trigger unauthorised behaviour across sessions.

Key facts. Cited by 6 papers in the map (data through 14 Oct 2026). Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use. Builds on 5 of its 16 references that are in the map. First posted on arXiv 14 Jul 2025.

Builds on

5 of 16 references.

References · 5

  1. 2022Constitutional AI: Harmlessness from AI Feedbackcited by 3,050
  2. 2022Red Teaming Language Models to Reduce Harms: Methods, Scaling Behaviors, and Lessons Learnedcited by 729
  3. 2024CyberSecEval 2: A Wide-Ranging Cybersecurity Evaluation Suite for Large Language Modelscited by 89
  4. 2025System Prompt Poisoning: Persistent Attacks on Large Language Models Beyond User Injectioncited by 14
  5. 2024Trust No AI: Prompt Injection Along The CIA Security Triadcited by 8

Built on by

Cited by year (2026 to 14 Oct)

Citing papers per year: 2025: 5; 2026: 1.

6 papers.

  1. 2025QSAF: A Novel Mitigation Framework for Cognitive Degradation in Agentic AIcited by 6
  2. 2025Fortifying the Agentic Web: A Unified Zero-Trust Architecture Against Logic-layer Threatscited by 3
  3. 2025AAGATE: A NIST AI RMF-Aligned Governance Platform for Agentic AIcited by 3
  4. 2025DIRF: A Framework for Digital Identity Protection and Clone Governance in Agentic AI Systemscited by 2
  5. 2025Empowering Real-World: A Survey on the Technology, Practice, and Evaluation of LLM-driven Industry Agentscited by 2
  6. 2026LAAF: Logic-layer Automated Attack Framework A Systematic Red-Teaming Methodology for LPCI Vulnerabilities in Agentic Large Language Model Systemscited by 1

Similar, not cited

Close in wording, no citation either way

  1. 2024Prompt Infection: LLM-to-LLM Prompt Injection within Multi-Agent Systems
  2. 2026Architecting Secure AI Agents: Perspectives on System-Level Defenses Against Indirect Prompt Injection Attacks
  3. 2024Signed-Prompt: A New Approach to Prevent Prompt Injection Attacks Against LLM-Integrated Applications
  4. 2025Preventing Prompt Injection with Type-Directed Privilege Separation
  5. 2026The Landscape of Prompt Injection Threats in LLM Agents: From Taxonomy to Analysis
  6. 2026A Layered Security Framework Against Prompt Injection in RAG-Based Chatbots
  7. 2025Prompt Flow Integrity to Prevent Privilege Escalation in LLM Agents
  8. 2026Skill-Inject: Measuring Agent Vulnerability to Skill File Attacks