Agon: An Autonomous Large-Scale Omnidisciplinary Research System Built on Prompt Economy

2026-06-23

AI Summary

While current large language models can efficiently generate scientific content, verifying the truthfulness of their claims has emerged as a critical bottleneck. This work proposes Agon, a fully prompt-driven, code-free autonomous research system designed to operate across all scientific disciplines. At its core lies the Prompt Economy mechanism, guided by six design principles—Prompt Economy, future-oriented reasoning, minimal prompting, universal disciplinary applicability, massive parallelism, and zero-code implementation—which enables the system to automatically validate testable assertions within its workflow, delegating only undecidable questions to human judgment. Over 444 iterative cycles, Agon demonstrated cross-domain autonomous research capability, confirming its scalability while uncovering novel failure modes and introducing a taxonomy that distinguishes between issues amenable to automatic resolution and those requiring human intervention.

Abstract

Large language models are making research production scalable, shifting the bottleneck from producing artifacts to judging claims. We present \textsc{Agon}, a research orchestrator that validates what can be checked inside the workflow and leaves the remaining judgments to human scientists. \textsc{Agon} is built on six design principles: Prompt Economy, Future-Facing, Minimal Prompts, OmniDisciplinary, Massive Parallelism, and Zero-Code. We ran \textsc{Agon} across domains for 444 iterations of Prompt Economy loops, using only small starting topics and no human-written experimental code. These deployments demonstrate scalability while exposing new classes of failure. We organize these failures into a taxonomy along severity, fixability, visibility, and capability locus. The taxonomy separates failures the loops can see and fix from those that require human judgment. Together, these results show that \textsc{Agon} is pushing research toward a new paradigm: machine scales, human steers.

Problem

• research validation
• claim judgment
• autonomous research system
• human-AI collaboration
• failure taxonomy

Innovation

• Prompt Economy
• Autonomous Research System
• Omnidisciplinary
• Zero-Code
• Failure Taxonomy