HASTE: Evolving Agent Harnesses Against Emerging Attacks Using Sparse Evidence
cs.AI
Submitted: 2026-10-02
Updated: 2026-10-02
Code: https://github.com/xxiqiao/HASTE
Sources
- Phi-4-Mini Technical Report: Compact yet Powerful Multimodal Language Models via Mixture-of-LoRAs
- ShieldAgent: Shielding Agents via Verifiable Safety Policy Reasoning
- LlamaFirewall: An open source guardrail system for building secure AI agents
- Defeating Prompt Injections by Design
- Promptbreeder: Self-Referential Self-Improvement Via Prompt Evolution
- FinHarness: An Inline Lifecycle Safety Harness for Finance LLM Agents
- SHADE-Arena: Evaluating Sabotage and Monitoring in LLM Agents
- Meta-Harness: End-to-End Optimization of Model Harnesses
- STAC: When Innocent Tools Form Dangerous Chains for LLM Agents
- Plant, Persist, Trigger: Sleeper Attack on Large Language Model Agents
- AutoFlow: Automated Workflow Generation for Large Language Model Agents
- SafeHarness: Lifecycle-Integrated Security Architecture for LLM-based Agent Deployment
- AgentDoG: A Diagnostic Guardrail Framework for AI Agent Safety and Security
- Prompt Injection attack against LLM-integrated Applications
- AutoHarness: improving LLM agents by automatically synthesizing a code harness
- Natural-Language Agent Harnesses
- Herculean: An Agentic Benchmark for Financial Intelligence
- Ignore Previous Prompt: Attack Techniques For Language Models
- Prompt Injection Attack to Tool Selection in LLM Agents
- MemoryGraft: Persistent Compromise of LLM Agents via Poisoned Experience Retrieval
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