Agent-integrated software improves how users control shared intelligent systems
Agent-Integrated Software: Interaction Contracts and Continuous Assurance
Software EngineeringArtificial Intelligence
Summary
It can be hard to manage smart agents that keep working inside apps while users change goals or data. The authors introduce Agent-Integrated Software, a way to create software that clearly links user tasks and app actions. They propose a model and rules to help users understand and control the agent's work continuously, ensuring its actions stay trustworthy. This approach aims to make integrating intelligent agents into software easier to build and maintain.
What this means in practice
- •For software developers: Build applications with embedded agents that users can understand and control reliably over time using interaction contracts.
- •For human-computer interaction teams: Design user interfaces that expose task-level semantics to supervise and adjust ongoing agent tasks within software.
A position paper. It proposes an approach and reports no results.
Authors
Shengcheng Yu, Chunrong Fang, Zhenyu Chen
Abstract
Embedding an intelligent agent in an existing application creates a persistent coordination problem: users can revise goals and manipulate shared objects while delegated execution continues. We argue that dependable integration requires an explicit correspondence between task-level interaction and application behavior. We introduce Agent-Integrated Software (AIS) as a software pattern combining a conventional core, direct interaction, and a built-in agent, and Intent-Level Interaction Abstraction (IIA) as the task semantics through which users inspect and control delegated work. An open transition-system model relates AIS execution to IIA states and events. Interaction contracts constrain this relation through task bindings, role-specific authority, control transitions, and outcome evidence; continuous assurance maintains scoped claims as their dependencies change. A compact disclosure contract and conditional propositions illustrate why local component validity is insufficient and how selected admission invariants can be separated from planning. Contrasting software domains expose the framework's assumptions and limits. This perspective develops a research agenda spanning application abstraction, development support, controlled execution, quality assessment, and human supervision, with the aim of making agent integration a maintainable software engineering discipline.