SafeCoEvo: Co-Evolving Safety Harnesses and Guards for LLM Agents at Test-Time
cs.AI
Submitted: 2026-09-29
Updated: 2026-10-02
Code: https://github.com/SII-YUCHENG2002/SafeCoEvo
Sources
- Constitutional AI: Harmlessness from AI Feedback
- HarnessX: A Composable, Adaptive, and Evolvable Agent Harness Foundry
- Llama Guard: LLM-based Input-Output Safeguard for Human-AI Conversations
- AgentDyn: Are Your Agent Security Defenses Deployable in Real-World Dynamic Environments?
- EvoSafeHarness: Evolving Model- and Domain-Specific Harnesses for Securing Agents
- SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment
- AgentDoG 1.5: A Lightweight and Scalable Alignment Framework for AI Agent Safety and Security
- AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security
- SafeEvolve: Harness-Policy Co-Evolution from Agent Experience for Safety Alignment
- SHE: Trajectory-driven Safety Harness Evolution for LLM Agents
- Progent: Securing AI Agents with Privilege Control
- SingGuard: A Policy-Adaptive Multimodal LLM Guardrail with Dynamic Reasoning
- A-MemGuard: A Proactive Defense Framework for LLM-Based Agent Memory
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