SAFETY SENTRY: Context-Aware Human Intervention via EXECUTE-ASK-REFUSE Routing
cs.AI
Submitted: 2026-07-15
Updated: 2026-10-08
Code: https://github.com/safesentry/SAFETY-SENTRY
Terminology
Sources
- STaR-GATE: Teaching Language Models to Ask Clarifying Questions
- Memory Injection Attacks on LLM Agents via Query-Only Interaction
- Towards Verifiably Safe Tool Use for LLM Agents
- TrustAgent: Towards Safe and Trustworthy LLM-based Agents
- Building a Foundational Guardrail for General Agentic Systems via Synthetic Data
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- ATBench: A Diverse and Realistic Agent Trajectory Benchmark for Safety Evaluation and Diagnosis
- AGrail: A Lifelong Agent Guardrail with Effective and Adaptive Safety Detection
- Fully Autonomous AI Agents Should Not be Developed
- ToolSafe: Enhancing Tool Invocation Safety of LLM-based agents via Proactive Step-level Guardrail and Feedback
- Agent Safety Alignment via Reinforcement Learning
- CRiskEval: A Chinese Multi-Level Risk Evaluation Benchmark Dataset for Large Language Models
- A Survey on Autonomy-Induced Security Risks in Large Model-Based Agents
- Structured Uncertainty guided Clarification for LLM Agents
- OpenAgentSafety: A Comprehensive Framework for Evaluating Real-World AI Agent Safety
- AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
- GuardAgent: Safeguard LLM Agents by a Guard Agent via Knowledge-Enabled Reasoning
- ShieldGemma: Generative AI Content Moderation Based on Gemma
- Agent-SafetyBench: Evaluating the Safety of LLM Agents
- Risky-Bench: Evaluating Agentic Safety Risks in Real-World Life-Assist Deployments
Related papers
- MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
- Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
- Agents' Last Exam
- The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
- MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
- Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems