Policy Compiler for Secure Agentic Systems

• URL: http://arxiv.org/abs/2602.16708v2
• Date: Thu, 19 Feb 2026 03:40:05 GMT
• Title: Policy Compiler for Secure Agentic Systems
• Authors: Nils Palumbo, Sarthak Choudhary, Jihye Choi, Prasad Chalasani, Somesh Jha,
• Abstract summary: We present PCAS, a Policy Compiler for Agentic Systems that provides deterministic policy enforcement.<n>We evaluate PCAS on three case studies: information flow policies for prompt injection defense, approval in a multi-agent pharmacovigilance system, and organizational policies for customer service.
• Score: 20.346157626726725
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: LLM-based agents are increasingly being deployed in contexts requiring complex authorization policies: customer service protocols, approval workflows, data access restrictions, and regulatory compliance. Embedding these policies in prompts provides no enforcement guarantees. We present PCAS, a Policy Compiler for Agentic Systems that provides deterministic policy enforcement. Enforcing such policies requires tracking information flow across agents, which linear message histories cannot capture. Instead, PCAS models the agentic system state as a dependency graph capturing causal relationships among events such as tool calls, tool results, and messages. Policies are expressed in a Datalog-derived language, as declarative rules that account for transitive information flow and cross-agent provenance. A reference monitor intercepts all actions and blocks violations before execution, providing deterministic enforcement independent of model reasoning. PCAS takes an existing agent implementation and a policy specification, and compiles them into an instrumented system that is policy-compliant by construction, with no security-specific restructuring required. We evaluate PCAS on three case studies: information flow policies for prompt injection defense, approval workflows in a multi-agent pharmacovigilance system, and organizational policies for customer service. On customer service tasks, PCAS improves policy compliance from 48% to 93% across frontier models, with zero policy violations in instrumented runs.

Related papers

• Taming Various Privilege Escalation in LLM-Based Agent Systems: A Mandatory Access Control Framework [16.14469140816631]
  Large Language Model (LLM)-based agent systems are increasingly deployed for complex real-world tasks.<n>This paper aims to understand and mitigate such attacks through the lens of privilege escalation.<n>We propose SEAgent, a mandatory access control framework built upon attribute-based access control (ABAC)<n>Our evaluations show that SEAgent effectively blocks various privilege escalation while maintaining a low false positive rate and negligible system overhead.
  arXiv  Detail & Related papers  (2026-01-17T03:22:56Z)
• Towards Verifiably Safe Tool Use for LLM Agents [53.55621104327779]
  Large language model (LLM)-based AI agents extend capabilities by enabling access to tools such as data sources, APIs, search engines, code sandboxes, and even other agents.<n>LLMs may invoke unintended tool interactions and introduce risks, such as leaking sensitive data or overwriting critical records.<n>Current approaches to mitigate these risks, such as model-based safeguards, enhance agents' reliability but cannot guarantee system safety.
  arXiv  Detail & Related papers  (2026-01-12T21:31:38Z)
• Securing the Model Context Protocol (MCP): Risks, Controls, and Governance [1.4072883206858737]
  We focus on three types of adversaries that take advantage of MCP s flexibility.<n>Based on early incidents and proof-of-concept attacks, we describe how MCP can increase the attack surface.<n>We propose a set of practical controls, including per-user authentication with scoped authorization.
  arXiv  Detail & Related papers  (2025-11-25T23:24:26Z)
• Policy Cards: Machine-Readable Runtime Governance for Autonomous AI Agents [0.0]
  Policy Cards are a machine-readable, deployment-layer standard for expressing operational, regulatory, and ethical constraints for AI agents.<n>Each Policy Card can be validated automatically, version-controlled, and linked to runtime enforcement or continuous-audit pipelines.
  arXiv  Detail & Related papers  (2025-10-28T12:59:55Z)
• Analyzing and Internalizing Complex Policy Documents for LLM Agents [53.14898416858099]
  Large Language Model (LLM)-based agentic systems rely on in-context policy documents encoding diverse business rules.<n>This motivates developing internalization methods that embed policy documents into model priors while preserving performance.<n>We introduce CC-Gen, an agentic benchmark generator with Controllable Complexity across four levels.
  arXiv  Detail & Related papers  (2025-10-13T16:30:07Z)
• Policy-as-Prompt: Turning AI Governance Rules into Guardrails for AI Agents [0.19336815376402716]
  We introduce a regulatory machine learning framework that converts unstructured design artifacts (like PRDs, TDDs, and code) into verifiable runtime guardrails.<n>Our Policy as Prompt method reads these documents and risk controls to build a source-linked policy tree.<n>System is built to enforce least privilege and data minimization.
  arXiv  Detail & Related papers  (2025-09-28T17:36:52Z)
• Few-shot Policy (de)composition in Conversational Question Answering [54.259440408606515]
  We propose a neuro-symbolic framework to detect policy compliance using large language models (LLMs) in a few-shot setting.<n>We show that our approach soundly reasons about policy compliance conversations by extracting sub-questions to be answered, assigning truth values from contextual information, and explicitly producing a set of logic statements from the given policies.<n>We apply this approach to the popular PCD and conversational machine reading benchmark, ShARC, and show competitive performance with no task-specific finetuning.
  arXiv  Detail & Related papers  (2025-01-20T08:40:15Z)
• Towards an Enforceable GDPR Specification [49.1574468325115]
  Privacy by Design (PbD) is prescribed by modern privacy regulations such as the EU's. One emerging technique to realize PbD is enforcement (RE) We present a set of requirements and an iterative methodology for creating formal specifications of legal provisions.
  arXiv  Detail & Related papers  (2024-02-27T09:38:51Z)
• Distributed-Training-and-Execution Multi-Agent Reinforcement Learning for Power Control in HetNet [48.96004919910818]
  We propose a multi-agent deep reinforcement learning (MADRL) based power control scheme for the HetNet. To promote cooperation among agents, we develop a penalty-based Q learning (PQL) algorithm for MADRL systems. In this way, an agent's policy can be learned by other agents more easily, resulting in a more efficient collaboration process.
  arXiv  Detail & Related papers  (2022-12-15T17:01:56Z)
• Mutual Information Regularized Offline Reinforcement Learning [76.05299071490913]
  We propose a novel MISA framework to approach offline RL from the perspective of Mutual Information between States and Actions in the dataset. We show that optimizing this lower bound is equivalent to maximizing the likelihood of a one-step improved policy on the offline dataset. We introduce 3 different variants of MISA, and empirically demonstrate that tighter mutual information lower bound gives better offline RL performance.
  arXiv  Detail & Related papers  (2022-10-14T03:22:43Z)
• Coordinating Policies Among Multiple Agents via an Intelligent Communication Channel [81.39444892747512]
  In Multi-Agent Reinforcement Learning (MARL), specialized channels are often introduced that allow agents to communicate directly with one another. We propose an alternative approach whereby agents communicate through an intelligent facilitator that learns to sift through and interpret signals provided by all agents to improve the agents' collective performance.
  arXiv  Detail & Related papers  (2022-05-21T14:11:33Z)

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.