mlsanityData to 7 Oct 2026

ST-WebAgentBench: A Benchmark for Evaluating Safety and Trustworthiness in Web Agents

arXiv 2410.06703 ↗First posted 9 Oct 2024cs.AIcited by 126

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

Abstract (arXiv)

Autonomous web agents solve complex browsing tasks, yet existing benchmarks measure only whether an agent finishes a task, ignoring whether it does so safely or in a way enterprises can trust. To integrate these agents into critical workflows, safety and trustworthiness (ST) are prerequisite conditions for adoption. We introduce textbftextscST-WebAgentBench, a configurable and easily extensible suite for evaluating web agent ST across realistic enterprise scenarios. Each of its 222 tasks is paired with ST policies, concise rules that encode constraints, and is scored along six orthogonal dimensions (e.g., user consent, robustness). Beyond raw task success, we propose the textitCompletion Under Policy (textitCuP) metric, which credits only completions that respect all applicable policies, and the textitRisk Ratio, which quantifies ST breaches across dimensions. Evaluating three open state-of-the-art agents reveals that their average CuP is less than two-thirds of their nominal completion rate, exposing critical safety gaps. By releasing code, evaluation templates, and a policy-authoring interface, hrefhttps://sites.google.com/view/st-webagentbench/hometextscST-WebAgentBench provides an actionable first step toward deploying trustworthy web agents at scale.

Key facts. Cited by 126 papers in the map (data through 7 Oct 2026). Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use. Builds on 42 of its 82 references that are in the map. First posted on arXiv 9 Oct 2024.

Builds on

42 of 82 references.

Closest ancestors · 4

Same direction, newest first

  1. 2024Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents3 Oct
  2. 2024GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning13 Jun
  3. 2024R-Judge: Benchmarking Safety Risk Awareness for LLM Agents18 Jan
  4. 2023Evil Geniuses: Delving into the Safety of LLM-based Agents20 Nov

Other references · 38

  1. 2022Survey of Hallucination in Natural Language Generationcited by 2,056
  2. 2023WebArena: A Realistic Web Environment for Building Autonomous Agentscited by 1,908
  3. 2023The Rise and Potential of Large Language Model Based Agents: A Surveycited by 1,422
  4. 2023AgentBench: Evaluating LLMs as Agentscited by 1,211
  5. 2023Mind2Web: Towards a Generalist Agent for the Webcited by 1,210
  6. 2022WebShop: Towards Scalable Real-World Web Interaction with Grounded Language Agentscited by 1,197
  7. 2024OpenHands: An Open Platform for AI Software Developers as Generalist Agentscited by 956
  8. 2023SelfCheckGPT: Zero-Resource Black-Box Hallucination Detection for Generative Large Language Modelscited by 787
  9. 2024SeeClick: Harnessing GUI Grounding for Advanced Visual GUI Agentscited by 553
  10. 2024GPT-4V(ision) is a Generalist Web Agent, if Groundedcited by 547
  11. 2024WebVoyager: Building an End-to-End Web Agent with Large Multimodal Modelscited by 424
  12. 2023Trustworthy LLMs: a Survey and Guideline for Evaluating Large Language Models' Alignmentcited by 394
26 more
  1. 2024WorkArena: How Capable Are Web Agents at Solving Common Knowledge Work Tasks?cited by 348
  2. 2024Agent Workflow Memorycited by 342
  3. 2018Reinforcement Learning on Web Interfaces Using Workflow-Guided Explorationcited by 282
  4. 2024TrustLLM: Trustworthiness in Large Language Modelscited by 280
  5. 2024Agent Q: Advanced Reasoning and Learning for Autonomous AI Agentscited by 201
  6. 2024AI Agents That Mattercited by 199
  7. 2024WebLINX: Real-World Website Navigation with Multi-Turn Dialoguecited by 177
  8. 2022A Review of Safe Reinforcement Learning: Methods, Theory and Applicationscited by 169
  9. 2024AssistantBench: Can Web Agents Solve Realistic and Time-Consuming Tasks?cited by 155
  10. 2024AutoWebGLM: A Large Language Model-based Web Navigating Agentcited by 130
  11. 2024WebCanvas: Benchmarking Web Agents in Online Environmentscited by 119
  12. 2023Clarify When Necessary: Resolving Ambiguity Through Interaction with LMscited by 93
  13. 2024WorkArena++: Towards Compositional Planning and Reasoning-based Common Knowledge Work Taskscited by 81
  14. 2024Agent-E: From Autonomous Web Navigation to Foundational Design Principles in Agentic Systemscited by 79
  15. 2024Safeguarding Large Language Models: A Surveycited by 73
  16. 2023SteP: Stacked LLM Policies for Web Actionscited by 66
  17. 2023Connecting the Dots in Trustworthy Artificial Intelligence: From AI Principles, Ethics, and Key Requirements to Responsible AI Systems and Regulationcited by 60
  18. 2024Introducing v0.5 of the AI Safety Benchmark from MLCommonscited by 60
  19. 2021Grounding Open-Domain Instructions to Automate Web Support Taskscited by 43
  20. 2024Tools Fail: Detecting Silent Errors in Faulty Toolscited by 23
  21. 2023Enhancing Trust in LLM-Based AI Automation Agents: New Considerations and Future Challengescited by 19
  22. 2024WebSuite: Systematically Evaluating Why Web Agents Failcited by 16
  23. 2024From Grounding to Planning: Benchmarking Bottlenecks in Web Agentscited by 12
  24. 2024SNAP: Semantic Stories for Next Activity Predictioncited by 8
  25. 2022Prescriptive Process Monitoring in Intelligent Process Automation with Chatbot Orchestrationcited by 6
  26. 2024IDA: Breaking Barriers in No-code UI Automation Through Large Language Models and Human-Centric Designcited by 5

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2024: 3; 2025: 49; 2026: 74.

126 papers.

Sort:

By area:

  1. 2025Survey on Evaluation of LLM-based Agentscited by 190
  2. 2024Large Language Model-Brained GUI Agents: A Surveycited by 160
  3. 2025A Survey of Self-Evolving Agents: What, When, How, and Where to Evolve on the Path to Artificial Super Intelligencecited by 155
  4. 2024The BrowserGym Ecosystem for Web Agent Researchcited by 152
  5. 2025ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoningcited by 144
  6. 2024GUI Agents: A Surveycited by 101
  7. 2025SafeArena: Evaluating the Safety of Autonomous Web Agentscited by 97
  8. 2025A Survey of WebAgents: Towards Next-Generation AI Agents for Web Automation with Large Foundation Modelscited by 87
  9. 2025OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safetycited by 71
  10. 2025Safety at Scale: A Comprehensive Survey of Large Model and Agent Safetycited by 63
  11. 2025REAL: Benchmarking Autonomous Agents on Deterministic Simulations of Real Websitescited by 43
  12. 2025Agentic Web: Weaving the Next Web with AI Agentscited by 43
38 more
  1. 2025RedTeamCUA: Realistic Adversarial Testing of Computer-Use Agents in Hybrid Web-OS Environmentscited by 42
  2. 2025UFO2: The Desktop AgentOScited by 38
  3. 2025OS Agents: A Survey on MLLM-based Agents for General Computing Devices Usecited by 35
  4. 2025RiOSWorld: Benchmarking the Risk of Multimodal Computer-Use Agentscited by 34
  5. 2025Towards Trustworthy GUI Agents: A Surveycited by 33
  6. 2025Towards Enterprise-Ready Computer Using Generalist Agentcited by 31
  7. 2025DoomArena: A framework for Testing AI Agents Against Evolving Security Threatscited by 31
  8. 2025Enforcing Temporal Constraints for LLM Agentscited by 31
  9. 2026ClawsBench: Evaluating Capability and Safety of LLM Productivity Agents in Simulated Workspacescited by 31
  10. 2025BEARCUBS: A benchmark for computer-using web agentscited by 29
  11. 2026OSWorld 2.0: Benchmarking Computer Use Agents on Long-Horizon Real-World Taskscited by 29
  12. 2025LLM-Based Human-Agent Collaboration and Interaction Systems: A Surveycited by 28
  13. 2025Agentic AI Security: Threats, Defenses, Evaluation, and Open Challengescited by 26
  14. 2025Symbiotic Cooperation for Web Agents: Harnessing Complementary Strengths of Large and Small LLMscited by 24
  15. 2025A Survey on the Safety and Security Threats of Computer-Using Agents: JARVIS or Ultron?cited by 23
  16. 2025MLA-Trust: Benchmarking Trustworthiness of Multimodal LLM Agents in GUI Environmentscited by 23
  17. 2025WebGuard: Building a Generalizable Guardrail for Web Agentscited by 20
  18. 2025OS-Sentinel: Towards Safety-Enhanced Mobile GUI Agents via Hybrid Validation in Realistic Workflowscited by 18
  19. 2025Build the web for agents, not agents for the webcited by 17
  20. 2025Towards Enforcing Company Policy Adherence in Agentic Workflowscited by 17
  21. 2025SecureWebArena: A Holistic Security Evaluation Benchmark for LVLM-based Web Agentscited by 17
  22. 2026Uncertainty Quantification in LLM Agents: Foundations, Emerging Challenges, and Opportunitiescited by 15
  23. 2026Beyond Task Completion: Revealing Corrupt Success in LLM Agents through Procedure-Aware Evaluationcited by 15
  24. 2025Think Twice Before You Act: Enhancing Agent Behavioral Safety with Thought Correctioncited by 14
  25. 2025Alignment and Safety in Large Language Models: Safety Mechanisms, Training Paradigms, and Emerging Challengescited by 14
  26. 2025WAREX: Web Agent Reliability Evaluation on Existing Benchmarkscited by 14
  27. 2026VIGIL: Defending LLM Agents Against Tool Stream Injection via Verify-Before-Commitcited by 13
  28. 2025Evolutionary Perspectives on the Evaluation of LLM-Based AI Agents: A Comprehensive Surveycited by 12
  29. 2025GhostEI-Bench: Do Mobile Agents Resilience to Environmental Injection in Dynamic On-Device Environments?cited by 12
  30. 2026The Evolution of Tool Use in LLM Agents: From Single-Tool Call to Multi-Tool Orchestrationcited by 11
  31. 2026Claw-Eval-Live: A Live Agent Benchmark for Evolving Real-World Workflowscited by 11
  32. 2025Effective Red-Teaming of Policy-Adherent Agentscited by 10
  33. 2025HarmonyGuard: Toward Safety and Utility in Web Agents via Adaptive Policy Enhancement and Dual-Objective Optimizationcited by 10
  34. 2026World of Workflows: A Benchmark for Bringing World Models to Enterprise Systemscited by 10
  35. 2025MLE-Smith: Scaling MLE Tasks with Automated Multi-Agent Pipelinecited by 9
  36. 2026BalDRO: A Distributionally Robust Optimization based Framework for Large Language Model Unlearningcited by 8
  37. 2026Beyond Accuracy: Policy Invariance as a Reliability Test for LLM Safety Judgescited by 8
  38. 2026Log analysis is necessary for credible evaluation of AI agentscited by 8

50 most cited of 126 All 126 (JSON)

Similar, not cited

Close in wording, no citation either way

  1. 2026Real-Time Trust Verification for Safe Agentic Actions using TrustBench
  2. 2026Training Needs Trustworthy Worlds: Verified Synthetic Web Environments for Agent Learning
  3. 2026WebTrap Park: An Automated Platform for Systematic Security Evaluation of Web Agents
  4. 2026Tencent WorkBuddy Bench: A Multi-Domain Coding-Agent Benchmark with Contamination-Resistant Task Construction
  5. 2026Towards trustworthy agentic AI: a comprehensive survey of safety, robustness, privacy, and system security
  6. 2026Beyond Benchmark Islands: Toward Representative Trustworthiness Evaluation for Agentic AI
  7. 2026OSGuard: A Benchmark for Safety in Computer-Use Agents
  8. 2026Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agents