Dataset · Hugging Face

ZJUICSR/ActBench

Name: ActBench
Creator: ZJUICSR
License: mit
Keywords: task_categories:text-generation, language:en, license:mit, size_categories:10K<n<100K, format:parquet, modality:text, library:datasets, library:dask, library:polars, library:mlcroissant, arxiv:2608.09476, region:us, benchmark, agents, agent-safety

ActBench ActBench is a benchmark for evaluating behavioral safety in tool-using cowork agents from execution trajectories. This dataset accompanies ActBench: Self-Evolving Benchmark of Behavioral Safety in Cowork Agents and the ZJUICSR/ActBench source repository. Dataset contents This release contains two configurations: task_pairs: 300 rows. Each row pairs one benign task with its adversarial variant and embeds the complete public UTF-8 task bundles: YAML… See the full description on the dataset page: https://huggingface.co/datasets/ZJUICSR/ActBench.

View on Hugging FaceReport an issue

Author / organization	ZJUICSR
License (as declared)	mit
Downloads	220 (as reported by Hugging Face)
Likes	4
Created	13 August 2026
Last updated at source	10 October 2026

benchmarkagentsagent-safetybehavioral-safetyprompt-injectionred-teamingtrajectories

Embed this entry

<iframe src="https://silytics.si/embed/item/31b79d95-3c57-4c61-adbd-5833b7f71ead" width="420" height="190" style="border:0" loading="lazy" title="Silytics entry"></iframe>

Free, no tracking. More about feeds and embeds.

More from ZJUICSR

ZJUICSR/DeepTrapDataset

Silytics shows public metadata from Hugging Face and links to the original. Information may be incomplete or out of date. Last refreshed 16 October 2026. Licenses and terms of the original apply to the work itself. Not financial, legal or professional advice.