EvoUndo: Recoverability-Constrained Self-Evolution for LLM Agent Harnesses

arXiv:2608.28363 · cs.AI · Submitted 2026-08-28 · Read on arXiv

cs.AI

Submitted: 2026-08-28

Updated: 2026-09-16

Code: https://github.com/cordiverse/paper

Terminology

• Verification — 27× in this paper · explained in 2 episodes
• Grounding — 24× in this paper · explained in 3 episodes
• Benchmark — 15× in this paper · explained in 2 episodes
• Verifier — 15× in this paper · explained in 2 episodes
• Trace — 14× in this paper · explained in 2 episodes
• Robustness — 5× in this paper · explained in 5 episodes
• Agent Harness — 3× in this paper · explained in 1 episode
• Self-Evolving Agents — 3× in this paper · explained in 2 episodes
• Failure Modes — 2× in this paper · explained in 3 episodes
• Failure Taxonomy — 2× in this paper · explained in 2 episodes
• Formal Verification — 2× in this paper · explained in 2 episodes
• Adversarial Noise — 1× in this paper · explained in 1 episode
• Agent Skills — 1× in this paper · explained in 1 episode
• Consistency — 1× in this paper · explained in 3 episodes
• Generalization — 1× in this paper · explained in 13 episodes
• Generative AI — 1× in this paper · explained in 2 episodes
• Scaffolding — 1× in this paper · explained in 3 episodes

Sources

• Tool-R0: Self-Evolving LLM Agents for Tool-Learning from Zero Data
• SEVerA: Verified Synthesis of Self-Evolving Agents
• Co-Harness: Co-Evolving Harnesses and Model Weights for LLM Agents
• Towards Formal Verification of LLM-Generated Code from Natural Language Prompts
• Autonomous Action Runtime Management(AARM):A System Specification for Securing AI-Driven Actions at Runtime
• Reversible effects as inverse arrows
• VIGIL: Runtime Enforcement of Behavioral Specifications in AI Agent Skills
• Agentic Harness Engineering: Observability-Driven Automatic Evolution of Coding-Agent Harnesses
• Evolving Agents in the Dark: Retrospective Harness Optimization via Self-Preference
• GoEX: Perspectives and Designs Towards a Runtime for Autonomous LLM Applications
• Goedel Machines: Self-Referential Universal Problem Solvers Making Provably Optimal Self-Improvements
• Harness-R1: Learning to Edit Executable Runtime Harnesses from Agent Failure Trajectories
• AgentSpec: Customizable Runtime Enforcement for Safe and Reliable LLM Agents
• EvoTool: Self-Evolving Tool-Use Policy Optimization in LLM Agents via Blame-Aware Mutation and Diversity-Aware Selection
• Zombie Agents: Persistent Control of Self-Evolving LLM Agents via Self-Reinforcing Injections
• G\"odel Agent: A Self-Referential Agent Framework for Recursive Self-Improvement
• Shepherd: Enabling Programmable Meta-Agents via Reversible Agentic Execution Traces
• Self-Harness: Harnesses That Improve Themselves
• Symbolic Learning Enables Self-Evolving Agents

Related papers

• MAVEN-T: Reinforced Heterogeneous Distillation for Real-Time Multi-Agent Trajectory Prediction
• Model Discovery Agent: LLM-assisted Bayesian experiment design for data-efficient discovery of mechanistic world models
• The Clinician's Veto: Navigating Trust, Liability, and Uncertainty in Autonomous AI Prescribing
• MindHelper: Closed-Loop Embodied Mental-State Reasoning for Precision Intervention
• Incumbent Advantage: Brand Bias and Cognitive Manipulation Dynamics in LLM Recommendation Systems
• VSAL: A Vision Solver with Adaptive Layouts for Graph Property Detection