Agent device foundation improves reliable execution across smart devices
ADF-EA: A Unified Execution Assurance System for Agent Device Foundation
Multiagent Systems
Summary
Many AI agents talk to different smart devices, but sometimes commands don’t work as expected or it’s unclear if they did. The authors present a system that helps AI agents understand what devices can do, check if actions succeed, and handle problems safely. This system keeps track of progress and allows retries or fixes when needed. Tests show it reduces mistakes and unnecessary actions, making smart devices work more reliably with AI.
What this means in practice
- •For robotics developers: Coordinate AI-driven robots with diverse tools to ensure commands complete reliably and recover safely from errors.
- •For home automation engineers: Build AI agents that control varied smart home devices with verified effects and managed retries when failures occur.
Authors
Xuechun Li, Jiaxin Liang, Jie Li, baolong Li, Jue Wang, Peng Yuan, Hang Huang
Abstract
Agents based on large language models (LLMs) can access heterogeneous devices through tools and APIs, but reliable execution must account for unmet effects, uncertain outcomes, and changing prerequisites. A command may be acknowledged without producing its intended effect, while missing feedback may obscure an action that has already succeeded. We present Agent Device Foundation--Execution Assurance (ADF-EA), an architecture that connects agent planning and device execution through shared capability contracts. Device Capability Contracts (DCCs) unify invocation conditions, intended effects, evidence requirements, and recovery rules across heterogeneous interfaces. Agents use these contracts to plan, while the runtime applies the same semantics to authorize actions, verify effects, and govern continuation and completion. Persistent execution state retains verified progress, unresolved outcomes, and remaining budgets across plan revisions, enabling observation-based recovery, authorized retries, and necessary state repair. We formalize the execution lifecycle and establish conditional soundness properties for completion and recovery authorization. Evaluations span multiple LLMs, five agent frameworks, and simulated process-control, household, and robotic manipulation domains. Compared with direct invocation and existing execution-checking approaches, ADF-EA reduces false completion and unnecessary repetition, supports necessary state repair, prevents calls to unavailable capabilities, and preserves permitted task completion and recovery. These results demonstrate DCCs as a reusable semantic foundation for agent autonomy across heterogeneous devices, unifying capability-based planning, evidence-grounded execution, and authorized recovery within one architecture.