Meta Uses DTap to Red-Team Muse Spark 1.3
DTap provides realistic environments and adaptive attacks for Meta’s tool-use and browser-use safety evaluations.
Read blog




































































































































































































A Controllable and Interactive
Red-Teaming Platform for AI Agents
The first dynamic red-teaming framework against AI Agents across 14 domains and 50+ sandboxed environments.
Covering Diverse Indirect Injections in Environments, Tools, Skills, and Direct Prompt Injections.
Paper available on arXiv
How teams use DTap to evaluate and secure AI agents.
Ranked by average attack success rate (higher = more vulnerable).
VulnerabilityRank | Agent | Model | Indirect ASR Lower = safer | Direct ASR Lower = safer | BSR Higher = more capable |
|---|---|---|---|---|---|
Each point is an agent. The bottom-right corner — high benign success, low attack success — is the ideal frontier. The dashed line traces the Pareto-optimal agents: nothing else is both safer and more capable.
Each environment is policy-aligned and spans real-world regulatory and operational constraints across our 14 supported domains.
If you use DecodingTrust-Agent in your research, please cite our paper.
@article{chen2026decodingtrust,
title={DecodingTrust-Agent Platform (DTap): A Controllable and Interactive Red-Teaming Platform for AI Agents},
author={Chen, Zhaorun and Liu, Xun and Tong, Haibo and Guo, Chengquan and Nie, Yuzhou and Zhang, Jiawei and Kang, Mintong and Xu, Chejian and Liu, Qichang and Liu, Xiaogeng and others},
journal={arXiv preprint arXiv:2605.04808},
year={2026}
}