Persona-Execution Separation: An Architecture Pattern for Evolving LLM Agents under Execution Audit

arXiv:2608.27427 · cs.SE, cs.AI · Submitted 2026-08-27 · Read on arXiv

cs.SE, cs.AI

Submitted: 2026-08-27

Updated: 2026-09-14

Comments: 36 pages

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Core — 15× in this paper · explained in 2 episodes
• Trace — 12× in this paper · explained in 2 episodes
• Decoupling — 6× in this paper · explained in 2 episodes
• Abstraction — 5× in this paper · explained in 1 episode
• Persona-Execution Separation — 5× in this paper
• Verification — 5× in this paper · explained in 2 episodes
• Affordance — 3× in this paper · explained in 1 episode
• Large Language Model (LLM) — 3× in this paper · explained in 3 episodes
• AI Agents — 2× in this paper · explained in 2 episodes
• Agent Architectures — 2× in this paper · explained in 1 episode
• Data Abstraction Firewall — 2× in this paper
• Generalization — 2× in this paper · explained in 13 episodes
• Model Context Protocol — 2× in this paper
• Proof-of-Execution Memory — 2× in this paper
• Adaptation — 1× in this paper · explained in 1 episode
• Belief Poisoning Attacks — 1× in this paper
• Calibration — 1× in this paper · explained in 7 episodes
• Communication Overhead — 1× in this paper · explained in 2 episodes
• Human-in-the-Loop — 1× in this paper · explained in 3 episodes
• International Conference on Learning Representations — 1× in this paper
• Multi-Agent System — 1× in this paper · explained in 6 episodes
• Provenance — 1× in this paper · explained in 4 episodes
• Xtensible Access Control Markup Language — 1× in this paper

Sources

• Firewalls to Secure Dynamic LLM Agentic Networks
• Securing AI Agents with Information-Flow Control
• Defeating Prompt Injections by Design
• SafeGPT: Preventing Data Leakage and Unethical Outputs in Enterprise LLM Use
• Harness-MU: A Safe, Governed, and Effective Harness for Multi-User LLM Agents
• AgentScope: A Flexible yet Robust Multi-Agent Platform
• Agentao: A Policy-Governed Runtime Harness for Embeddable Tool-Using LLM Agents
• Progressive Crystallization: Turning Agent Exploration into Deterministic, Lower-Cost Workflows in Production
• Proof-of-Execution Memory: Defending LLM Agents Against Forged-Reasoning Attacks by Verifying What Actually Happened
• Toolformer: Language Models Can Teach Themselves to Use Tools
• Scalable and Transferable Black-Box Jailbreaks for Language Models via Persona Modulation
• Progent: Securing AI Agents with Privilege Control
• Organizational Control Layer: Governance Infrastructure at the Execution Boundary of LLM Agent Systems
• AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
• When Agents See Humans as the Outgroup: Belief-Dependent Bias in LLM-Powered Agents
• AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation
• ReAct: Synergizing Reasoning and Acting in Language Models

Related papers

• Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
• GitSkills: A Dataset of Agent Skills on GitHub
• SABER: Benchmarking Operational Safety of LLM Coding Agents in Stateful Project Workspaces
• PackMonitor: Enabling Zero Package Hallucinations Through Decoding-Time Monitoring
• IntentCoding: Amplifying User Intent in Code Generation
• Incentives and Outcomes in Bug Bounties