# Prime Agent: A Self-Improving RLM Harness

arXiv: https://arxiv.org/abs/2608.23552 · HTML: https://mlsanity.com/paper/2608.23552/

## Key facts

- Cited by 13 papers in the map (data through 14 Oct 2026).
- Placed in Software engineering coding agents, one of 16 directions in LLM agents and tool use.
- Builds on 28 of its 38 references that are in the map.
- First posted on arXiv 24 Aug 2026.

## Placed in

- LLM agents and tool use / [Software engineering coding agents](https://mlsanity.com/agents/software-engineering-coding-agents/)
- Synthetic data and self-improvement for LLMs / [Self-evolving LLM agents](https://mlsanity.com/synth/self-evolving-llm-agents/)

## Builds on (most cited first, 28 of 28)

- [ReAct: Synergizing Reasoning and Acting in Language Models](https://mlsanity.com/paper/2210.03629/) (2022)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Toolformer: Language Models Can Teach Themselves to Use Tools](https://mlsanity.com/paper/2302.04761/) (2023)
- [Generative Agents: Interactive Simulacra of Human Behavior](https://mlsanity.com/paper/2304.03442/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [AutoGen: Enabling Next-Gen LLM Applications via Multi-Agent Conversation](https://mlsanity.com/paper/2308.08155/) (2023)
- [Scaling LLM Test-Time Compute Optimally can be More Effective than Scaling Model Parameters](https://mlsanity.com/paper/2408.03314/) (2024)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [OpenHands: An Open Platform for AI Software Developers as Generalist Agents](https://mlsanity.com/paper/2407.16741/) (2024)
- [STaR: Bootstrapping Reasoning With Reasoning](https://mlsanity.com/paper/2203.14465/) (2022)
- [Executable Code Actions Elicit Better LLM Agents](https://mlsanity.com/paper/2402.01030/) (2024)
- [LongBench v2: Towards Deeper Understanding and Reasoning on Realistic Long-context Multitasks](https://mlsanity.com/paper/2412.15204/) (2024)
- [Training Software Engineering Agents and Verifiers with SWE-Gym](https://mlsanity.com/paper/2412.21139/) (2024)
- [Theory of Mind for Multi-Agent Collaboration via Large Language Models](https://mlsanity.com/paper/2310.10701/) (2023)
- [AgentTrek: Agent Trajectory Synthesis via Guiding Replay with Web Tutorials](https://mlsanity.com/paper/2412.09605/) (2024)
- [Continual Harness: Online Adaptation for Self-Improving Foundation Agents](https://mlsanity.com/paper/2605.09998/) (2026)
- [Oolong: Evaluating Long Context Reasoning and Aggregation Capabilities](https://mlsanity.com/paper/2511.02817/) (2025)
- [LongBench Pro: A More Realistic and Comprehensive Bilingual Long-Context Evaluation Benchmark](https://mlsanity.com/paper/2601.02872/) (2026)
- [Executable World Models for ARC-AGI-3 in the Era of Coding Agents](https://mlsanity.com/paper/2605.05138/) (2026)
- [PRO-LONG: Programmatic Memory Enables Long-Horizon Reasoning](https://mlsanity.com/paper/2607.20064/) (2026)
- [OBLIQ-Bench: Exposing Overlooked Bottlenecks in Modern Retrievers with Latent and Implicit Queries](https://mlsanity.com/paper/2605.06235/) (2026)
- [OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3](https://mlsanity.com/paper/2607.01531/) (2026)
- [Interpretable Learned Emergent Communication for Human-Agent Teams](https://mlsanity.com/paper/2201.07452/) (2022)
- [Many-Tier Instruction Hierarchy in LLM Agents](https://mlsanity.com/paper/2604.09443/) (2026)
- [Towards True Lossless Sparse Communication in Multi-Agent Systems](https://mlsanity.com/paper/2212.00115/) (2022)
- [On the Role of Emergent Communication for Social Learning in Multi-Agent Reinforcement Learning](https://mlsanity.com/paper/2302.14276/) (2023)

## Built on by (most cited first, 13 of 13)

- [RSIAgent: Autonomous Exploration for Recursive Self-improvement in New Environments](https://mlsanity.com/paper/2609.15364/) (2026)
- [Harness-Zero: Harness Distillation via Agent-as-Harness](https://mlsanity.com/paper/2609.24974/) (2026)
- [Mid-Harness: Scaling Actions Between Model and Harness for Terminal Agents](https://mlsanity.com/paper/2609.39982/) (2026)
- [StudentBench: AI and human tutoring yield equivalent GRE learning gains](https://mlsanity.com/paper/2609.28470/) (2026)
- [Witness: Discovery, Deciphering, and Epiphany in Interactive Puzzle Environments](https://mlsanity.com/paper/2609.32208/) (2026)
- [Schema: Discovering Unknown Environments via Agentic Program Induction](https://mlsanity.com/paper/2609.39140/) (2026)
- [Trust Is Not a Score: Runtime Assurance Contracts for High-Risk AI Agents](https://mlsanity.com/paper/2609.39717/) (2026)
- [Kepler: Auditable World Models for ARC-AGI-3](https://mlsanity.com/paper/2610.00834/) (2026)
- [Wikidata Search Traces: A Dataset for Training Knowledge Graph Search Agents](https://mlsanity.com/paper/2610.06650/) (2026)
- [ServeLearnBench: How Well Can Agents Self-Improve from Serving Experience?](https://mlsanity.com/paper/2610.07792/) (2026)
- [LongHarness: Advancing Agentic Reasoning over 100M-Token Corpora via Stateful Evidence Management](https://mlsanity.com/paper/2610.13075/) (2026)
- [From Existing Methods to New Research Ideas: Two Perspectives on Recursive Self-Improvement](https://mlsanity.com/paper/2610.14625/) (2026)
- [ProgAgent: Learning Tool Orchestration through Programmatic Training](https://mlsanity.com/paper/2610.14774/) (2026)

## Abstract (arXiv)

Language models are sequential processors, but long-horizon agency requires external information and computation beyond model weights and active context. Prime Agent is an open-source harness for long-horizon evaluation and coding-agent workflows. A persistent IPython REPL follows the Recursive Language Model abstraction for programmatic context processing and test-time compute, while Continual Harness preserves histories, memories, skills, prompts, and subagent specifications across trajectories. Recursive subagents coordinate through direct agent-to-agent communication, and the Agents View lets humans inspect and manage daemon-backed sessions. Prime Agent standardizes execution, recovery, verification, and resource accounting while leaving strategy construction to the model. This low-friction, expressive membrane prevents harness failures from becoming model failures and pushes measurement toward the model's true maximal underlying capability. Prime Agent raises ARC-AGI-3 RHAE Best@1 from 30% to 95.5% and matches or exceeds native and popular harnesses across long-context coding, GPU-kernel generation, emulator construction, and autonomous nanoGPT speedruns. On Factorio, we find refinement allows for continuous technology progression and dedicated subagents enable parallelized work. Code is available at https://github.com/PrimeIntellect-ai/prime-agent.