Paper Number

ICIS2025-1870

Paper Type

Short

Abstract

This paper presents an Action Design Research (ADR) study on the Agentic Office System (AOS)—a multi-agent, LLM-integrated workspace developed to address fragmentation and coordination challenges in modern knowledge work. AOS unifies specialized AI agents through a shared memory layer, natural language interface, and structured collaboration mechanisms, guided by six user Requirements (R1–R6) and seven Design Principles (DP1–DP7). The system introduces the Consensus Canvas as a focal interaction model, enabling users and agents to collaboratively scope, align, and manage tasks within a unified workspace. Drawing on iterative prototyping and user engagements, AOS demonstrates how agentic systems can reduce tool switching, improve task efficiency, and strengthen transparency and oversight. AOS contributes both a working prototype and practical insights into how multi-agent, AI-augmented systems can enhance productivity and collaboration by supporting adaptive, transparent, and user-aligned workflows in enterprise environments.

Comments

13-DesignDevPM

Download

Share

COinS
 
Dec 14th, 12:00 AM

An ADR Approach on a Multi-Agent System for Team Productivity and Collaboration

This paper presents an Action Design Research (ADR) study on the Agentic Office System (AOS)—a multi-agent, LLM-integrated workspace developed to address fragmentation and coordination challenges in modern knowledge work. AOS unifies specialized AI agents through a shared memory layer, natural language interface, and structured collaboration mechanisms, guided by six user Requirements (R1–R6) and seven Design Principles (DP1–DP7). The system introduces the Consensus Canvas as a focal interaction model, enabling users and agents to collaboratively scope, align, and manage tasks within a unified workspace. Drawing on iterative prototyping and user engagements, AOS demonstrates how agentic systems can reduce tool switching, improve task efficiency, and strengthen transparency and oversight. AOS contributes both a working prototype and practical insights into how multi-agent, AI-augmented systems can enhance productivity and collaboration by supporting adaptive, transparent, and user-aligned workflows in enterprise environments.

When commenting on articles, please be friendly, welcoming, respectful and abide by the AIS eLibrary Discussion Thread Code of Conduct posted here.