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

arXiv: https://arxiv.org/abs/2606.24177 · HTML: https://mlsanity.com/paper/2606.24177/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 20 of its 26 references that are in the map.
- First posted on arXiv 23 Jun 2026.

## Builds on (most cited first, 20 of 20)

- [The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery](https://mlsanity.com/paper/2408.06292/) (2024)
- [Agent Laboratory: Using LLM Agents as Research Assistants](https://mlsanity.com/paper/2501.04227/) (2025)
- [The AI Scientist-v2: Workshop-Level Automated Scientific Discovery via Agentic Tree Search](https://mlsanity.com/paper/2504.08066/) (2025)
- [PaperBench: Evaluating AI's Ability to Replicate AI Research](https://mlsanity.com/paper/2504.01848/) (2025)
- [CycleResearcher: Improving Automated Research via Automated Review](https://mlsanity.com/paper/2411.00816/) (2024)
- [MLR-Bench: Evaluating AI Agents on Open-Ended Machine Learning Research](https://mlsanity.com/paper/2505.19955/) (2025)
- [Can LLM feedback enhance review quality? A randomized study of 20K reviews at ICLR 2025](https://mlsanity.com/paper/2504.09737/) (2025)
- [InternAgent: When Agent Becomes the Scientist -- Building Closed-Loop System from Hypothesis to Verification](https://mlsanity.com/paper/2505.16938/) (2025)
- [When AI Co-Scientists Fail: SPOT-a Benchmark for Automated Verification of Scientific Research](https://mlsanity.com/paper/2505.11855/) (2025)
- [ScientistOne: Towards Human-Level Autonomous Research via Chain-of-Evidence](https://mlsanity.com/paper/2605.26340/) (2026)
- [AgentV-RL: Scaling Reward Modeling with Agentic Verifier](https://mlsanity.com/paper/2604.16004/) (2026)
- [SoundnessBench: Can Your AI Scientist Really Tell Good Research Ideas from Bad Ones?](https://mlsanity.com/paper/2605.30329/) (2026)
- [BadScientist: Can a Research Agent Write Convincing but Unsound Papers that Fool LLM Reviewers?](https://mlsanity.com/paper/2510.18003/) (2025)
- [From Fluent to Verifiable: Claim-Level Auditability for Deep Research Agents](https://mlsanity.com/paper/2602.13855/) (2026)
- [Tree-of-Debate: Multi-Persona Debate Trees Elicit Critical Thinking for Scientific Comparative Analysis](https://mlsanity.com/paper/2502.14767/) (2025)
- [Evidence-Bound Autonomous Research (EviBound): A Governance Framework for Eliminating False Claims](https://mlsanity.com/paper/2511.05524/) (2025)
- [DeployBench: Benchmarking LLM Agents for Research Artifact Deployment](https://mlsanity.com/paper/2606.05238/) (2026)
- [PerspectiveGap: A Benchmark for Multi-Agent Orchestration Prompting](https://mlsanity.com/paper/2606.08878/) (2026)
- [Sibyl-AutoResearch: Autonomous Research Needs Self-Evolving Trial-and-Error Harnesses, Not Paper Generators](https://mlsanity.com/paper/2605.22343/) (2026)
- [NORA: A Harness-Engineered Autonomous Research Agent for End-to-End Spatial Data Science](https://mlsanity.com/paper/2605.02092/) (2026)

## Built on by (most cited first, 4 of 4)

- [Bringing Agentic Search to Earth Observation Data Discovery](https://mlsanity.com/paper/2607.02387/) (2026)
- [Understanding Sparse Attention Selectivity in Long-Context Foundation Models via Counterfactual Evaluation](https://mlsanity.com/paper/2608.01676/) (2026)
- [AgonAlpha: Autonomous Alpha Discovery via Prompt Economy and Scalable Agentic Search](https://mlsanity.com/paper/2608.11250/) (2026)
- [AutoSR: Automatic Symbolic Regression by Searching Research States](https://mlsanity.com/paper/2608.16876/) (2026)

## Abstract (arXiv)

Large language models are making research production scalable, shifting the bottleneck from producing artifacts to judging claims. We present textscAgon, a research orchestrator that validates what can be checked inside the workflow and leaves the remaining judgments to human scientists. textscAgon is built on six design principles: Prompt Economy, Future-Facing, Minimal Prompts, OmniDisciplinary, Massive Parallelism, and Zero-Code. We ran textscAgon 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 textscAgon is pushing research toward a new paradigm: machine scales, human steers.