Safety
Sapien: A Stateful Policy Engine for Autonomous AI Agents
arXiv:2610.00797v1 Announce Type: new Abstract: Contextual security defenses prevent AI agents from taking rogue actions by synthesizing a task-specific policy and enforcing it on the agent's tool cal
arXiv:2610.00797v1 Announce Type: new Abstract: Contextual security defenses prevent AI agents from taking rogue actions by synthesizing a task-specific policy and enforcing it on the agent's tool calls. In multi-step tasks, however, which actions are valid often depends on what the agent has already done and learned. We present Sapien, a policy engine for enforcing stateful contextual policies. A Sapien policy specifies permitted tool-call sequences using a regular expression extended with stateful predicates, deferred policy generation, and scoped semantic checks. We show that Sapien stays within a few percent of an unconstrained agent's utility. Even if the agent is fully hijacked, Sapien's policies rule out 93-95% of attacks on AgentDojo and 62-85% on Toolathlon (twice as many as tool allowlists on long-horizon tasks).
Related
- ActionGuard: Tool Call Authorization under Poisoned Skills
- ActGov: Governing LLM Agent Actions via Policy-Constrained Validation
- AIRGuard: Guarding Agent Actions with Runtime Authority Control
- Symbolic Guardrails for Domain-Specific Agents: Stronger Safety and Security Guarantees Without Sacrificing Utility
- Securing Multi-Tool AI Agent Chains With Dynamic, Real-Time Compositional Policies
Source: arXiv cs.AI | 2026-10-02