ADL: A Declarative Language for Agent-Based Chatbots

2025-04-21

AI Summary

Non-technical customer service personnel face significant challenges in constructing task-oriented multi-agent chatbots due to the complexity of traditional programming and prompt engineering. Method: This paper proposes ADL (Agent Definition Language), a natural-language-driven, declarative agent programming paradigm designed for business users. ADL enables specification of agent roles, interaction logic, and tool integrations entirely in natural language, abstracting away execution details while uniformly supporting custom functions, external API calls, and third-party agent collaboration. We design MICA, a dedicated interpreter, alongside a plug-in–based tool integration framework to parse and execute ADL programs. Contribution/Results: The open-source MICA system substantially lowers the barriers to prompt engineering and debugging. Empirical evaluation shows that domain experts can build production-grade customer service bots within hours—achieving, for the first time, a programming-free, reusable, and extensible methodology for intelligent agent system construction.

Technology Category

• Multiagent Systems: Agent Communication
• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Cognitive Modeling & Cognitive Systems: Agent Architectures

Application Category

• Semantics and Knowledge: Data modeling to support human-machine intelligence, including LLMs agents, intelligent system behavior, explanations, and user-friendly interactions
• Search and Retrieval-Augmented AI: Agentic search
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

There are numerous agent frameworks capable of creating and orchestrating agents to address complex tasks. However, these frameworks are often too complicated for customer service professionals, who may not have much programming experience but still need an easy way to create chatbots with rich business logic. In this work, we introduce ADL, a Declarative Language for Agent-Based Chatbots. ADL simplifies chatbot development by using natural language programming at its core, making it easier for a broad audience to customize and build task-oriented chatbots. It includes four types of agents and supports integration with custom functions, tool use, and third-party agents. ADL abstracts away implementation details, offering a declarative way to define agents and their interactions, which could ease prompt engineering, testing and debugging. MICA, a multi-agent system designed to interpret and execute ADL programs, has been developed and is now available as an open-source project at https://github.com/Mica-labs/MICA. Its user documentation can be found at https://mica-labs.github.io/.

Problem

• Simplifies chatbot development for non-programmers
• Enables creation of task-oriented chatbots with rich logic
• Abstracts implementation details for easier agent customization

Innovation

• Declarative language simplifies chatbot development
• Natural language programming for broad accessibility
• Supports custom functions and third-party integrations