# OpenClawBench: Benchmarking Process-side Anomalies in Real-world Agent Execution Trajectories

arXiv: https://arxiv.org/abs/2605.29253 · HTML: https://mlsanity.com/paper/2605.29253/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 18 of its 25 references that are in the map.
- First posted on arXiv 28 May 2026.

## Builds on (most cited first, 18 of 18)

- [LoRA: Low-Rank Adaptation of Large Language Models](https://mlsanity.com/paper/2106.09685/) (2021)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [τ-bench: A Benchmark for Tool-Agent-User Interaction in Real-World Domains](https://mlsanity.com/paper/2406.12045/) (2024)
- [OSWorld: Benchmarking Multimodal Agents for Open-Ended Tasks in Real Computer Environments](https://mlsanity.com/paper/2404.07972/) (2024)
- [Agent Security Bench (ASB): Formalizing and Benchmarking Attacks and Defenses in LLM-based Agents](https://mlsanity.com/paper/2410.02644/) (2024)
- [Agent-SafetyBench: Evaluating the Safety of LLM Agents](https://mlsanity.com/paper/2412.14470/) (2024)
- [AgentAuditor: Human-Level Safety and Security Evaluation for LLM Agents](https://mlsanity.com/paper/2506.00641/) (2025)
- [ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback](https://mlsanity.com/paper/2601.10156/) (2026)
- [Claw-Eval: Towards Trustworthy Evaluation of Autonomous Agents](https://mlsanity.com/paper/2604.06132/) (2026)
- [AgentProcessBench: Diagnosing Step-Level Process Quality in Tool-Using Agents](https://mlsanity.com/paper/2603.14465/) (2026)
- [TrajAD: Trajectory Anomaly Detection for Trustworthy LLM Agents](https://mlsanity.com/paper/2602.06443/) (2026)
- [Wink: Recovering from Misbehaviors in Coding Agents](https://mlsanity.com/paper/2602.17037/) (2026)
- [Cloud-OpsBench: A Reproducible Benchmark for Agentic Root Cause Analysis in Cloud Systems](https://mlsanity.com/paper/2603.00468/) (2026)
- [Auditable Agents](https://mlsanity.com/paper/2604.05485/) (2026)
- [HINTBench: Horizon-agent Intrinsic Non-attack Trajectory Benchmark](https://mlsanity.com/paper/2604.13954/) (2026)
- [Harnessing Embodied Agents: Runtime Governance for Policy-Constrained Execution](https://mlsanity.com/paper/2604.07833/) (2026)
- [Verify Before You Commit: Towards Faithful Reasoning in LLM Agents via Self-Auditing](https://mlsanity.com/paper/2604.08401/) (2026)

## Built on by (most cited first, 4 of 4)

- [Rethinking Self-Evolving Agent Skills: Feedback Dynamics over Multiple Rounds](https://mlsanity.com/paper/2608.02636/) (2026)
- [Graph Engineering in the Era of LLM Agents: From Individual Intelligence to System Intelligence](https://mlsanity.com/paper/2608.21156/) (2026)
- [ClawTrack: Towards Trace-Level Evaluation and Improvement of Real-World Autonomous Agents](https://mlsanity.com/paper/2607.28037/) (2026)
- [TraceDance: An Automated System for Building Agent Behavior Benchmarks from Real-World Agent Deployment Traces](https://mlsanity.com/paper/2609.33295/) (2026)

## Abstract (arXiv)

Task success can hide process anomalies in real-world agent executions. An agent may pass the final task oracle while still accumulating unresolved ambiguity, unsafe external writes, ignored errors, weakly grounded commitments, or capability-boundary overcommitment. We study this mismatch as the Outcome-Process Gap and introduce OpenClawBench, a large-scale dataset for measuring and supervising process-side anomalies in real agent execution processes. OpenClawBench is built from BFCL-driven OpenClaw sessions produced by 6 source models and contains 31,264 annotated trajectories. It aligns task-oracle outcomes with structured process evidence. FullTax converts the aligned trajectories into structured anomaly supervision: binary labels, supporting evidence, onset/span localization, severity, recoverability, and a 5-class anomaly taxonomy. Using OpenClawBench, we make the Outcome-Process Gap measurable. Among 31,135 oracle-passing executions, 2,904 are still labeled process-anomalous under FullTax. These results show that success-only evaluation misses a concrete class of process-side failures in real agent executions. A LoRA-fine-tuned Gemma 3 12B detector trained on the high-confidence FullTax supervised pool reaches binary F1=0.729 on the cleaner-labels held-out test split. Together, OpenClawBench turns real agent execution logs into auditable and reusable supervision for studying, diagnosing, and operationally monitoring runtime agent reliability.