PluginRSI: Recursive Improvement of Agent Harnesses with Reusable Plugins

2026-09-26

AI Summary

This work addresses the difficulty of isolating and reusing independent mechanisms within existing agent frameworks. We propose a recursive self-improvement paradigm based on reusable plugins that decouples frameworks into atomic components, enabling the independent evolution and efficient composition of mechanisms through iterative optimization, accumulation, and reorganization of a plugin library. The core methodology encompasses a plugin-based architectural design, shared library management, a recursive self-improvement algorithm, and cross-model transfer techniques. Experimental results demonstrate that the proposed approach significantly outperforms existing baselines across multiple tasks. Furthermore, the accumulated plugin library effectively accelerates subsequent optimization processes and exhibits strong cross-model generalization capabilities.

Abstract

The harness surrounding a language model is a central determinant of agent performance. Recent methods optimize harnesses by searching over complete programs, where individual mechanisms are difficult to isolate and reuse. We introduce PluginRSI, which represents a harness as a composition of atomized plugins and organizes harness evolution around these plugins. Individual plugins are improved independently and accumulated in a shared library, then recombined into new harnesses at each iteration. PluginRSI improves over existing harness optimization methods across software engineering, command-line interaction, and question-answering tasks. The resulting harnesses retain their advantage when transferred to other solver models without further optimization. The evolved plugin library accelerates subsequent optimization from the initial harness, which helps faster and higher convergence on unseen tasks. These results show that accumulating reusable mechanisms provides an effective basis for continued harness improvement.

Problem

• Agent Harness
• Harness Optimization
• Plugin Reusability
• Recursive Improvement
• Language Model Agents

Innovation

• PluginRSI
• Reusable Plugins
• Harness Optimization
• Recursive Improvement
• Agent Framework