# BraveGuard: From Open-World Threats to Safer Computer-Use Agents

arXiv: https://arxiv.org/abs/2606.01166 · HTML: https://mlsanity.com/paper/2606.01166/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 37 of its 44 references that are in the map.
- First posted on arXiv 31 May 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 30 of 37)

- [The Llama 3 Herd of Models](https://mlsanity.com/paper/2407.21783/) (2024)
- [Qwen3 Technical Report](https://mlsanity.com/paper/2505.09388/) (2025)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations](https://mlsanity.com/paper/2312.06674/) (2023)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [Prompt Injection attack against LLM-integrated Applications](https://mlsanity.com/paper/2306.05499/) (2023)
- [AgentPoison: Red-teaming LLM Agents via Poisoning Memory or Knowledge Bases](https://mlsanity.com/paper/2407.12784/) (2024)
- [Kimi-VL Technical Report](https://mlsanity.com/paper/2504.07491/) (2025)
- [NeMo Guardrails: A Toolkit for Controllable and Safe LLM Applications with Programmable Rails](https://mlsanity.com/paper/2310.10501/) (2023)
- [R-Judge: Benchmarking Safety Risk Awareness for LLM Agents](https://mlsanity.com/paper/2401.10019/) (2024)
- [PoisonedRAG: Knowledge Corruption Attacks to Retrieval-Augmented Generation of Large Language Models](https://mlsanity.com/paper/2402.07867/) (2024)
- [Qwen3.5-Omni Technical Report](https://mlsanity.com/paper/2604.15804/) (2026)
- [Qwen3Guard Technical Report](https://mlsanity.com/paper/2510.14276/) (2025)
- [OpenClaw-RL: Train Any Agent Simply by Talking](https://mlsanity.com/paper/2603.10165/) (2026)
- [TrojanRAG: Retrieval-Augmented Generation Can Be Backdoor Driver in Large Language Models](https://mlsanity.com/paper/2405.13401/) (2024)
- [AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security](https://mlsanity.com/paper/2601.18491/) (2026)
- [Dive into Claude Code: The Design Space of Today's and Future AI Agent Systems](https://mlsanity.com/paper/2604.14228/) (2026)
- [Let It Flow: Agentic Crafting on Rock and Roll, Building the ROME Model within an Open Agentic Learning Ecosystem](https://mlsanity.com/paper/2512.24873/) (2025)
- [Don't Let the Claw Grip Your Hand: A Security Analysis and Defense Framework for OpenClaw](https://mlsanity.com/paper/2603.10387/) (2026)
- [Taming OpenClaw: Security Analysis and Mitigation of Autonomous LLM Agent Threats](https://mlsanity.com/paper/2603.11619/) (2026)
- [BadSkill: Backdoor Attacks on Agent Skills via Model-in-Skill Poisoning](https://mlsanity.com/paper/2604.09378/) (2026)
- [SkillTrojan: Backdoor Attacks on Skill-Based Agent Systems](https://mlsanity.com/paper/2604.06811/) (2026)
- [YuFeng-XGuard: A Reasoning-Centric, Interpretable, and Flexible Guardrail Model for Large Language Models](https://mlsanity.com/paper/2601.15588/) (2026)
- [LoCoBench-Agent: An Interactive Benchmark for LLM Agents in Long-Context Software Engineering](https://mlsanity.com/paper/2511.13998/) (2025)
- [BackdoorAgent: A Unified Framework for Backdoor Attacks on LLM-based Agents](https://mlsanity.com/paper/2601.04566/) (2026)
- [A Trajectory-Based Safety Audit of Clawdbot (OpenClaw)](https://mlsanity.com/paper/2602.14364/) (2026)
- [TraceSafe: A Systematic Assessment of LLM Guardrails on Multi-Step Tool-Calling Trajectories](https://mlsanity.com/paper/2604.07223/) (2026)
- [Clawdrain: Exploiting Tool-Calling Chains for Stealthy Token Exhaustion in OpenClaw Agents](https://mlsanity.com/paper/2603.00902/) (2026)
- [A Systematic Security Evaluation of OpenClaw and Its Variants](https://mlsanity.com/paper/2604.03131/) (2026)
- [Internal Safety Collapse in Frontier Large Language Models](https://mlsanity.com/paper/2603.23509/) (2026)

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

- [Safety Testing LLM Agents at Scale: From Risk Discovery to Evidence-Grounded Verification](https://mlsanity.com/paper/2607.01793/) (2026)
- [REDAgentBench: Executable Red Teaming and Faithful Measurement of LLM Agent Systems](https://mlsanity.com/paper/2608.10669/) (2026)
- [HazardAuditor: From Executable Threats to Safer Computer-Use Agents](https://mlsanity.com/paper/2609.15134/) (2026)
- [Trajectory Unlearning on LLM-based Agents](https://mlsanity.com/paper/2609.33639/) (2026)

## Abstract (arXiv)

Computer-use agents extend language models from text generation to sustained interaction with files, terminals, browsers, and external tools. This shift creates safety risks that are difficult to detect from isolated prompts or final responses, because harm often emerges only through multi-step execution traces whose individual actions appear locally benign. We introduce BraveGuard, a self-evolving defense framework for training guard models from open-world threat signals and realistic agent trajectories. BraveGuard mines recent research sources to identify emerging risks and attack patterns, instantiates them as executable computer-use tasks, collects agent rollouts, and derives trajectory-level supervision for guard model training. As new threats and validation failures appear, the pipeline can be repeated, yielding an adaptive defense loop rather than a static, benchmark-driven training process. We instantiate BraveGuard by training multiple guard backbones, including Qwen3-Guard and Llama-Guard variants, and evaluate the resulting guards on trajectory-level agent-safety benchmarks. BraveGuard consistently improves safety detection across computer-use trajectories. On AgentHazard, it substantially improves detection accuracy over off-the-shelf guard models, with accuracy increasing from 38.79% to 82.38% under the averaged guard-model setting. These results show that guard supervision grounded in open-world threat discovery and realistic agent execution can improve safety monitoring beyond fixed taxonomies and synthetic prompt-level data. BraveGuard offers a scalable path toward adaptive defenses for computer-use agents facing evolving real-world risks.