cs.CLNov 2, 2026

VeriHarm-Bench: Measuring Harmful Success in LLM Coding Agents Verifiably

Authors: Wei-Lin Chen, Zhepei Wei, Xinyu Zhu, Yu-Min Tseng, Tzu-Han Lin, Yu Meng

Organizations: University of Virginia · Virginia Tech · UCLA

Abstract

Large language model (LLM) coding agents increasingly operate with shell access, allowing them to write scripts, move files, alter permissions, and modify system configurations. However, an agent may achieve a user's goal while causing unintended harm, such as deleting unrelated files, weakening access controls, or corrupting data, in response to benign inputs without explicit jailbreaking. This failure mode is particularly important because apparent task success may reassure users that an agent behaved correctly, causing harmful side effects to go unnoticed. We call this harmful success: the agent successfully completes the intended benign task while simultaneously causing unintended harm. Despite its importance, harmful success is difficult to measure because conventional task-success metrics overlook harmful behaviors, while existing safety evaluations are often designed solely to assess harmful outcomes. We introduce VeriHarm, an automatic framework for constructing programmatically verifiable benchmarks that directly measure harmful success. VeriHarm minimally perturbs coding tasks to elicit unintended harmful behavior while maintaining the benignity of the task and preserving the original task objective. It then synthesizes executable harm tests that, together with the original task-correctness tests, evaluate both task accuracy and harmful outcomes. Frontier agents, including GPT-5.6-Sol and Claude-Opus-5, achieve an average task accuracy of 95% on our curated VeriHarm-Bench, yet still struggle to complete tasks safely: their safe success rate, the fraction of tasks completed successfully without triggering the designated harm, averages only 37%. These results suggest that successful task completion can frequently coexist with unintended harm, and that task-success metrics alone may substantially understate the potential safety risks of coding agents.

Figures & tables

Appendix figures & tables9 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Oct 26, 2026

V3^3AST-Bench: Verifiable, Valid, and Versatile Safety Evaluation for Coding Agents

Large language models increasingly act through code, modifying repositories, executing programs, and interacting with services in agent form. Their actions introduce risks that cannot be assessed from generated responses alone. Safety evaluation must verify actual consequences, distinguish harm avoidance from task incapability, and cover realistic workflows. We introduce V3^3AST-Bench, a benchmark for Verifiable results, Valid interpretation, and Versatile risk coverage of Agent Safety. It comprises 212 risk tasks spanning 18 categories. Dedicated runtime environments verify harm and task outcomes through executable checks over states, records, and artifact behavior, while paired capability twin tasks verify workflow competence. Our experiments span 51 combinations of 11 models and five harnesses, spanning 86,496 trials and over 9 billion tokens of model interaction. Results show that V3^3AST-Bench remains challenging even for frontier coding agents, as Claude Opus 4.8 with Claude Code still causes verified harm on 20.9% of tasks. Across model-harness combinations, agents with comparable competence exhibit markedly different safety outcomes, showing that task capability alone does not guarantee safe execution. More importantly, even explicit recognition of risks does not reliably prevent harmful actions, revealing a critical gap between risk awareness and safe execution. These findings highlight the need to move beyond risk recognition toward enforcing safety throughout execution.
Apr 3, 2026

AgentHazard: A Benchmark for Evaluating Harmful Behavior in Computer-Use Agents

Computer-use agents extend language models from text generation to persistent action over tools, files, and execution environments. Unlike chat systems, they maintain state across interactions and translate intermediate outputs into concrete actions. This creates a distinct safety challenge in that harmful behavior may emerge through sequences of individually plausible steps, including intermediate actions that appear locally acceptable but collectively lead to unauthorized actions. We present AgentHazard, a benchmark for evaluating harmful behavior in computer-use agents. AgentHazard contains 2,653 instances spanning diverse risk categories and attack strategies. Each instance pairs a harmful objective with a sequence of operational steps that are locally legitimate but jointly induce unsafe behavior. The benchmark evaluates whether agents can recognize and interrupt harm arising from accumulated context, repeated tool use, intermediate actions, and dependencies across steps. We evaluate AgentHazard on Claude Code, OpenClaw, and IFlow using mostly open or openly deployable models from the Qwen3, Kimi, GLM, and DeepSeek families. Our experimental results indicate that current systems remain highly vulnerable. In particular, when powered by Qwen3-Coder, Claude Code exhibits an attack success rate of 73.63%, suggesting that model alignment alone does not reliably guarantee the safety of autonomous agents.
Oct 1, 2025

Breaking the Code: Security Assessment of AI Code Agents Through Systematic Jailbreaking Attacks

Code-capable large language model (LLM) agents are embedded in software engineering workflows where they can read, write, and execute code, raising "jailbreak" stakes beyond text-only settings. Prior evaluations emphasize refusal or harmful-text detection, leaving open whether agents compile and run malicious programs. We present JAWS-Bench (Jailbreaks Across WorkSpaces), a benchmark spanning three escalating workspace regimes mirroring attacker capability: empty (JAWS-0), single-file (JAWS-1), and multi-file (JAWS-M). We pair it with a hierarchical, executable-aware Judge Framework that tests (i) compliance, (ii) attack success, (iii) syntactic correctness, and (iv) runtime executability to measure deployable harm. Across seven LLM backends from five families, prompt-only attacks in JAWS-0 achieve 61% compliance; 58% are harmful, 52% parse, and 27% run end-to-end. In JAWS-1, compliance reaches ~100% for stronger models with a mean ASR (Attack Success Rate) ~71%; JAWS-M raises mean ASR to ~75%, with 32% runnable attack code. Wrapping an LLM in an agent increases ASR by 1.6×\times, by overturning initial refusals during planning and tool use. Additional evaluations with SWE-Agent and OpenAI Codex exhibit similar trends, indicating that JAWS-Bench can be reused across multiple agent frameworks. Category analyses identify which attack classes are most vulnerable and deployable, motivating execution-aware defenses and refusal-preserving agent designs.