# SkillAxe: Sharpening LLM-Authored Agent Skills Through Evaluation-Guided Self-Refinement

arXiv: https://arxiv.org/abs/2606.10546 · HTML: https://mlsanity.com/paper/2606.10546/

## Key facts

- Cited by 12 papers in the map (data through 14 Oct 2026).
- Placed in Agent skill learning and evolution, one of 16 directions in LLM agents and tool use.
- Builds on 25 of its 33 references that are in the map.
- First posted on arXiv 9 Jun 2026.

## Placed in

- LLM agents and tool use / [Agent skill learning and evolution](https://mlsanity.com/agents/agent-skill-learning-and-evolution/)

## Builds on (most cited first, 25 of 25)

- [Judging LLM-as-a-Judge with MT-Bench and Chatbot Arena](https://mlsanity.com/paper/2306.05685/) (2023)
- [Reflexion: Language Agents with Verbal Reinforcement Learning](https://mlsanity.com/paper/2303.11366/) (2023)
- [Self-Refine: Iterative Refinement with Self-Feedback](https://mlsanity.com/paper/2303.17651/) (2023)
- [G-Eval: NLG Evaluation using GPT-4 with Better Human Alignment](https://mlsanity.com/paper/2303.16634/) (2023)
- [Voyager: An Open-Ended Embodied Agent with Large Language Models](https://mlsanity.com/paper/2305.16291/) (2023)
- [Large Language Models Are Human-Level Prompt Engineers](https://mlsanity.com/paper/2211.01910/) (2022)
- [Large Language Models as Optimizers](https://mlsanity.com/paper/2309.03409/) (2023)
- [SkillsBench: Benchmarking How Well Agent Skills Work Across Diverse Tasks](https://mlsanity.com/paper/2602.12670/) (2026)
- [Which Agent Causes Task Failures and When? On Automated Failure Attribution of LLM Multi-Agent Systems](https://mlsanity.com/paper/2505.00212/) (2025)
- [SoK: Agentic Skills -- Beyond Tool Use in LLM Agents](https://mlsanity.com/paper/2602.20867/) (2026)
- [Agent Skills for Large Language Models: Architecture, Acquisition, Security, and the Path Forward](https://mlsanity.com/paper/2602.12430/) (2026)
- [SpreadsheetBench: Towards Challenging Real World Spreadsheet Manipulation](https://mlsanity.com/paper/2406.14991/) (2024)
- [Reinforcement Learning for Self-Improving Agent with Skill Library](https://mlsanity.com/paper/2512.17102/) (2025)
- [Trace is the Next AutoDiff: Generative Optimization with Rich Feedback, Execution Traces, and LLMs](https://mlsanity.com/paper/2406.16218/) (2024)
- [How Well Do Agentic Skills Work in the Wild: Benchmarking LLM Skill Usage in Realistic Settings](https://mlsanity.com/paper/2604.04323/) (2026)
- [SkillRouter: Skill Routing for LLM Agents at Scale](https://mlsanity.com/paper/2603.22455/) (2026)
- [AgentRefine: Enhancing Agent Generalization through Refinement Tuning](https://mlsanity.com/paper/2501.01702/) (2025)
- [SkillNet: Create, Evaluate, and Connect AI Skills](https://mlsanity.com/paper/2603.04448/) (2026)
- [ProtoVAE: A Trustworthy Self-Explainable Prototypical Variational Model](https://mlsanity.com/paper/2210.08151/) (2022)
- [SkillReducer: Optimizing LLM Agent Skills for Token Efficiency](https://mlsanity.com/paper/2603.29919/) (2026)
- [Large Language Model Instruction Following: A Survey of Progresses and Challenges](https://mlsanity.com/paper/2303.10475/) (2023)
- [SkillCraft: Can LLM Agents Learn to Use Tools Skillfully?](https://mlsanity.com/paper/2603.00718/) (2026)
- [Align-Pro: A Principled Approach to Prompt Optimization for LLM Alignment](https://mlsanity.com/paper/2501.03486/) (2025)
- [SkillMOO: Multi-Objective Optimization of Agent Skills for Software Engineering](https://mlsanity.com/paper/2604.09297/) (2026)
- [Co-Evolving LLM Decision and Skill Bank Agents for Long-Horizon Tasks](https://mlsanity.com/paper/2604.20987/) (2026)

## Built on by (most cited first, 12 of 12)

- [Recursive Self-Improvement in AI: From Bounded Self-Refinement to Autonomous Research Loops](https://mlsanity.com/paper/2607.07663/) (2026)
- [Skill Self-Play: Pushing the Frontier of LLM Capability with Co-Evolving Skills](https://mlsanity.com/paper/2607.22529/) (2026)
- [SkillCoach: Self-Evolving Rubrics for Evaluating and Enhancing Agentic Skill-Use](https://mlsanity.com/paper/2607.01874/) (2026)
- [SkillAdam: Stable and Efficient Skill Evolution for Agents](https://mlsanity.com/paper/2609.08944/) (2026)
- [SkillAligner: Treating Retrieved Skills as Adaptable Drafts at Execution Time](https://mlsanity.com/paper/2608.06880/) (2026)
- [SkillShapley: Boundary-Adaptive Shapley Valuation for Skill Step Attribution in LLM Agents](https://mlsanity.com/paper/2608.13173/) (2026)
- [Safe Skill Retirement for Physical Agents](https://mlsanity.com/paper/2609.29543/) (2026)
- [Compression, structure, and executor capability: a controlled real-cost decomposition of language-model agent skill optimisation](https://mlsanity.com/paper/2607.03048/) (2026)
- [Self-Supervised Skill Optimization](https://mlsanity.com/paper/2607.28777/) (2026)
- [EVISKILL: Grounding Skill Evolution in Replayable Evidence](https://mlsanity.com/paper/2610.05030/) (2026)
- [SkillSandbox: Skill Verification via Dynamic Scenario Synthesis](https://mlsanity.com/paper/2610.10088/) (2026)
- [Trajectory-Guided Fault Localization for Agent Skill Evolution](https://mlsanity.com/paper/2610.11858/) (2026)

## Abstract (arXiv)

Skill documents, structured natural-language instructions that guide Large Language Model (LLM) agents, are critical to modern agent frameworks, yet LLMs struggle to write skills that actually work. On SkillsBench, human-authored skills improve pass rates by 16.2 percentage points, while LLM-authored skills provide no measurable gain. We introduce SkillAxe, a fully unsupervised framework that enables LLMs to iteratively diagnose and refine their own skills. SkillAxe decomposes skill quality into four interpretable dimensions (quality impact, trigger precision, instruction compliance with fault attribution, and solution-path coverage), producing structured improvement briefs that require no ground-truth labels, test suites, or environment rewards. On SkillsBench, SkillAxe improves pass rates by 28% relative over unimproved LLM skills and closes 47--67% of the gap to human-authored skills. We validate the approach as a continuous improvement engine in the wild on SpreadsheetBench, where a SkillAxe-built skill library learns from past agent trajectories and raises pass rate from 16.0% to 52.0% using only 22 skills.