Securing Web Agents with the Compile-Then-Execute Paradigm
cs.CR
Submitted: 2026-11-04
Updated: 2026-11-04
Code: https://github.com/apechow/Baya
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Sources
- Design Patterns for Securing LLM Agents against Prompt Injections
- DoomArena: A framework for Testing AI Agents Against Evolving Security Threats
- StruQ: Defending Against Prompt Injection with Structured Queries
- The BrowserGym Ecosystem for Web Agent Research
- Securing AI Agents with Information-Flow Control
- Defeating Prompt Injections by Design
- AgentDojo: A Dynamic Environment to Evaluate Prompt Injection Attacks and Defenses for LLM Agents
- DeepSeek-V4: Towards Highly Efficient Million-Token Context Intelligence
- WASP: Benchmarking Web Agent Security Against Prompt Injection Attacks
- CaMeLs Can Use Computers Too: System-level Security for Computer Use Agents
- OpAgent: Operator Agent for Web Navigation
- The Task Shield: Enforcing Task Alignment to Defend Against Indirect Prompt Injection in LLM Agents
- EIA: Environmental Injection Attack on Generalist Web Agents for Privacy Leakage
- ceLLMate: Sandboxing Browser AI Agents
- The Attacker Moves Second: Stronger Adaptive Attacks Bypass Defenses Against Llm Jailbreaks and Prompt Injections
- Untrusted Content Masking for Web Agents with Security Guarantees
- Jatmo: Prompt Injection Defense by Task-Specific Finetuning
- Web Agents Should Adopt the Compile-and-Execute Paradigm
- RestGPT: Connecting Large Language Models with Real-World RESTful APIs
- Beyond Browsing: API-Based Web Agents
Related papers
- From Strategy Search to Policy Learning: MCTS-Guided Adversarial Co-Evolution for LLM Safety Alignment
- Privacy Leakage from Recurrent State in LLM Inference
- Incrementally Verifiable Oracle Aided Computation
- Privilege and confidentiality in generative AI workflows
- BadMoE: Backdooring Mixture-of-Experts LLMs via Routing-Aware Triggers and Task-Underutilized Experts
- CoVault: Secure, Scalable Analytics of Personal Data