Code as Agent Harness

2026-05-18

AI Summary

This work proposes the “Code as Agent Suite” framework, elevating code from a passive output target to the foundational infrastructure of agent-based systems. Addressing limitations in existing AI systems—particularly regarding executability, verifiability, and state management—the framework integrates reasoning, memory, tool invocation, and multi-agent coordination through a three-layer architecture comprising interface, mechanism, and extension layers. Leveraging the code comprehension and generation capabilities of large language models, the system enables stateful, verifiable single- and multi-agent applications. It demonstrates high reliability and scalability across diverse domains, including GUI automation, scientific discovery, programming assistance, and DevOps, thereby establishing, for the first time, a systematic paradigm that positions code as the operational substrate for intelligent agents.

Abstract

Recent large language models (LLMs) have demonstrated strong capabilities in understanding and generating code, from competitive programming to repository-level software engineering. In emerging agentic systems, code is no longer only a target output. It increasingly serves as an operational substrate for agent reasoning, acting, environment modeling, and execution-based verification. We frame this shift through the lens of agent harnesses and introduce code as agent harness: a unified view that centers code as the basis for agent infrastructure. To systematically study this perspective, we organize the survey around three connected layers. First, we study the harness interface, where code connects agents to reasoning, action, and environment modeling. Second, we examine harness mechanisms: planning, memory, and tool use for long-horizon execution, together with feedback-driven control and optimization that make harness reliable and adaptive. Third, we discuss scaling the harness from single-agent systems to multi-agent settings, where shared code artifacts support multi-agent coordination, review, and verification. Across these layers, we summarize representative methods and practical applications of code as agent harness, spanning coding assistants, GUI/OS automation, embodied agents, scientific discovery, personalization and recommendation, DevOps, and enterprise workflows. We further outline open challenges for harness engineering, including evaluation beyond final task success, verification under incomplete feedback, regression-free harness improvement, consistent shared state across multiple agents, human oversight for safety-critical actions, and extensions to multimodal environments. By centering code as the harness of agentic AI, this survey provides a unified roadmap toward executable, verifiable, and stateful AI agent systems.

Problem

• code as agent harness
• agentic AI
• executable agents
• multi-agent coordination
• verification

Innovation

• code as agent harness
• agentic AI
• executable reasoning
• multi-agent coordination
• feedback-driven control