AgentEvolver: System-Wide Self-Evolution Through Task Execution

2026-10-08

AI Summary

This study addresses the challenge of transforming agent task experience into reusable capabilities by proposing a component-level self-evolution framework that operates over a fixed base model. The framework introduces eight categories of entities sharing a runtime and persistent planning mechanism, integrated with versioned lifecycle management and resumable context techniques to enable continuous capability accumulation and evolution. Experimental results demonstrate that the proposed method achieves an 82.08% resolution rate on SWE-bench Pro, effectively validating the feasibility of cross-scenario capability retention and reuse.

Abstract

An agent can complete a task without improving how it works. Turning task experience into reusable capability requires connecting the changed component to its evaluation and subsequent use. We present AgentEvolver, a system for developing capabilities during task execution while keeping the foundation model fixed. Eight entity families expose reusable operations, methods, agents, control flow, interfaces, and supporting state to revision through a common versioned lifecycle. A shared Runtime coordinates ongoing work, while persistent planning and recoverable context preserve task direction and supporting evidence. We evaluate task outcomes on SWE-bench Pro Public and examine capability changes in six application cases. The team reports an 82.08\% resolution rate with evolution, exceeding its reported baseline without evolution. The cases show retained capabilities entering later website, game, and research work, while also documenting incomplete objectives and an unsuccessful strategy. These findings distinguish improvement in a reusable component from success on the final task. AgentEvolver provides a concrete basis for studying capability accumulation through execution; independent-task transfer and total development cost remain open questions.

Problem

• Self-Evolution
• Capability Accumulation
• Task Execution
• Reusable Capability
• AI Agent

Innovation

• Self-Evolution
• Fixed Foundation Model
• Versioned Lifecycle
• Reusable Capabilities
• Shared Runtime