Twin Agent: Context Residual Compression for Privilege Separated Agents

• URL: http://arxiv.org/abs/2607.19595v1
• Date: Tue, 21 Jul 2026 21:47:52 GMT
• Title: Twin Agent: Context Residual Compression for Privilege Separated Agents
• Abstract summary: Large language model (LLM) agents are vulnerable to security risks, such as prompt injection attacks from untrusted context.<n>We propose Twin Agent, a general privilege separation design pattern inspired by residual coding in the agent context.<n>We evaluate Twin Agent on long-horizon software engineering tasks with SWE-bench Lite and on heterogeneous multi-tool interaction tasks with AgentDojo and DecodingTrust-Agent.
• Score: 18.474974915236245
• License: http://creativecommons.org/licenses/by/4.0/
• Abstract: Large language model (LLM) agents are vulnerable to security risks, such as prompt injection attacks from untrusted context that manipulate downstream reasoning and tool use. Existing secure-by-design approaches mitigate this risk by separating untrusted observations from privileged execution and careful control of information flow, but often degrade utility and require extensive task-specific engineering. We thus propose Twin Agent, a general privilege separation design pattern inspired by residual coding in the agent context. Twin Agent consists of two nearly symmetric agents: an Explore Agent that inspects untrusted information and a Safe Agent that executes privileged actions. The Explore Agent is conditioned on the Safe Agent's current context and communicates only compact hints to the Safe Agent about the next action to take. This design reduces the information needed to preserve task utility and thus achieves a better security--utility tradeoff, which we empirically verify by measuring how utility and attack success change as the length of hints varies. We evaluate Twin Agent on long-horizon software engineering tasks with SWE-bench Lite and on heterogeneous multi-tool interaction tasks with AgentDojo and DecodingTrust-Agent. Across both benchmarks, Twin Agent preserves high task utility while preventing prompt injection attacks, outperforming both undefended agents and privilege separation baselines.

Related papers

• AgentLens: Interpretable Safety Steering via Mechanistic Subspaces for Multi-Turn Coding Agent [53.82005364479556]
  Coding agents based on large language models (LLMs) demonstrate remarkable autonomous capabilities.<n>Existing safety mechanisms mainly rely on external guardrails.<n>We propose AgentLens, a white-box defense framework that performs runtime safety detection and representation-level mitigation.
  arXiv  Detail & Related papers  (2026-06-21T21:13:55Z)
• When Agents Handle Secrets: A Survey of Confidential Computing for Agentic AI [6.1511441479993225]
  Agentic AI systems that plan, invoke tools, maintain persistent memory, and delegate tasks to peer agents introduce a threat surface that differs materially from standalone model inference.<n> Confidential computing offers a hardware-rooted alternative: Trusted Execution Environments (TEEs) isolate agent code and data from privileged system software.<n>This survey synthesizes the design space in four parts: (i) a unified taxonomy of six TEE platforms covering deployment roles and performance tradeoffs; (ii) an agent-centric threat model spanning perception, planning, memory, action, and coordination layers mapped to nine security goals; (iii) a comparative survey of CC
  arXiv  Detail & Related papers  (2026-05-04T23:09:16Z)
• SafeAgent: A Runtime Protection Architecture for Agentic Systems [4.4767259565994495]
  This paper presents SafeAgent, a runtime security architecture that treats agent safety as a stateful decision problem.<n>The proposed design separates execution governance from semantic risk reasoning through two coordinated components.<n> Experiments on Agent Security Bench (ASB) and InjecAgent show that SafeAgent consistently improves robustness over baseline and text-level guardrail methods.
  arXiv  Detail & Related papers  (2026-04-19T18:02:21Z)
• AgentSentry: Mitigating Indirect Prompt Injection in LLM Agents via Temporal Causal Diagnostics and Context Purification [25.817251923574286]
  We propose a novel inference-time detection and mitigation framework for large language model (LLM) agents.<n>AgentSentry is the first inference-time defense to model multi-turn IPI as a temporal causal takeover.<n>We evaluate AgentSentry on the textscAgentDojo benchmark across four task suites, three IPI attack families, and multiple black-box LLMs.
  arXiv  Detail & Related papers  (2026-02-26T07:59:10Z)
• SkillJect: Automating Stealthy Skill-Based Prompt Injection for Coding Agents with Trace-Driven Closed-Loop Refinement [120.52289344734415]
  We propose an automated framework for stealthy prompt injection tailored to agent skills.<n>The framework forms a closed loop with three agents: an Attack Agent that synthesizes injection skills under explicit stealth constraints, a Code Agent that executes tasks using the injected skills and an Evaluate Agent that logs action traces.<n>Our method consistently achieves high attack success rates under realistic settings.
  arXiv  Detail & Related papers  (2026-02-15T16:09:48Z)
• OMNI-LEAK: Orchestrator Multi-Agent Network Induced Data Leakage [59.3826294523924]
  We investigate the security vulnerabilities of a popular multi-agent pattern known as the orchestrator setup.<n>We report the susceptibility of frontier models to different categories of attacks, finding that both reasoning and non-reasoning models are vulnerable.
  arXiv  Detail & Related papers  (2026-02-13T21:32:32Z)
• AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security [126.49733412191416]
  Current guardrail models lack agentic risk awareness and transparency in risk diagnosis.<n>We propose a unified three-dimensional taxonomy that categorizes agentic risks by their source (where), failure mode (how), and consequence (what)<n>We introduce a new fine-grained agentic safety benchmark (ATBench) and a Diagnostic Guardrail framework for agent safety and security (AgentDoG)
  arXiv  Detail & Related papers  (2026-01-26T13:45:41Z)
• Toward Trustworthy Agentic AI: A Multimodal Framework for Preventing Prompt Injection Attacks [0.0]
  This paper proposes a Cross-Agent Multimodal Provenanc- Aware Defense Framework for agentic AI systems.<n>The framework contains a Text sanitizer agent, visual sanitizer agent, and output validator agent all coordinated by a provenance ledger.<n>Experiments show that multimodal injection detection accuracy is significantly enhanced, and the cross-agent trust leakage is minimized.
  arXiv  Detail & Related papers  (2025-12-29T15:54:33Z)
• AgentArmor: Enforcing Program Analysis on Agent Runtime Trace to Defend Against Prompt Injection [14.522205401511727]
  Large Language Model (LLM) agents offer a powerful new paradigm for solving various problems by combining natural language reasoning with the execution of external tools.<n>In this work, we propose a novel insight that treats the agent runtime traces as structured programs with analyzable semantics.<n>We present AgentArmor, a program analysis framework that converts agent traces into graph intermediate representation-based structured program dependency representations.
  arXiv  Detail & Related papers  (2025-08-02T07:59:34Z)
• AgentVigil: Generic Black-Box Red-teaming for Indirect Prompt Injection against LLM Agents [54.29555239363013]
  We propose a generic black-box fuzzing framework, AgentVigil, to automatically discover and exploit indirect prompt injection vulnerabilities.<n>We evaluate AgentVigil on two public benchmarks, AgentDojo and VWA-adv, where it achieves 71% and 70% success rates against agents based on o3-mini and GPT-4o.<n>We apply our attacks in real-world environments, successfully misleading agents to navigate to arbitrary URLs, including malicious sites.
  arXiv  Detail & Related papers  (2025-05-09T07:40:17Z)
• On the Resilience of LLM-Based Multi-Agent Collaboration with Faulty Agents [58.79302663733703]
  Large language model-based multi-agent systems have shown great abilities across various tasks due to the collaboration of expert agents.<n>The impact of clumsy or even malicious agents--those who frequently make errors in their tasks--on the overall performance of the system remains underexplored.<n>This paper investigates what is the resilience of various system structures under faulty agents on different downstream tasks.
  arXiv  Detail & Related papers  (2024-08-02T03:25:20Z)
• GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning [79.07152553060601]
  We propose GuardAgent, the first guardrail agent to protect target agents by dynamically checking whether their actions satisfy given safety guard requests.<n>Specifically, GuardAgent first analyzes the safety guard requests to generate a task plan, and then maps this plan into guardrail code for execution.<n>We show that GuardAgent effectively moderates the violation actions for different types of agents on two benchmarks with over 98% and 83% guardrail accuracies, respectively.
  arXiv  Detail & Related papers  (2024-06-13T14:49:26Z)

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.