MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation

Hugging Face Daily Papers Papers
skill-centric agent-framework llm-agents skill-creation memory management evaluation

Summary

MUSE-Autoskill proposes a skill-centric agent framework that enables LLM agents to continuously create, reuse, and refine skills through a unified lifecycle of creation, memory, management, evaluation, and refinement. Experiments on SkillsBench show that lifecycle-managed skills improve task success, efficiency, reuse, and cross-agent transfer.

Large language model (LLM) agents rely on reusable skills to solve complex tasks. However, existing skill creation approaches treat skills as isolated and static artifacts, limiting their reusability, reliability, and long-term improvement. We propose MUSE-Autoskill Agent (Memory-Utilizing Skill Evolution), a skill-centric agent framework that lets agents continuously improve their task-solving capability by creating, reusing, and refining skills under a unified lifecycle (creation, memory, management, evaluation, and refinement). Our framework enables agents to create skills on demand, store and reuse them across tasks, organize and select them efficiently, and evaluate them through unit tests and runtime feedback for continuous refinement. We further introduce skill-level memory that accumulates experience for each skill across tasks, enabling more effective reuse and adaptation over time. Experiments on SkillsBench provide initial evidence that lifecycle-managed skills can improve task success, efficiency, reuse, and cross-agent transfer, highlighting the importance of treating skills as long-lived, experience-aware, and testable assets.
Original Article
View Cached Full Text

Cached at: 05/27/26, 02:47 AM

Paper page - MUSE-Autoskill: Self-Evolving Agents via Skill Creation, Memory, Management, and Evaluation

Source: https://huggingface.co/papers/2605.27366

Abstract

A skill-centric agent framework enables continuous improvement of task-solving capabilities through a unified lifecycle of skill creation, memory, management, evaluation, and refinement.

Large language model (LLM) agents rely on reusable skills to solve complex tasks. However, existingskill creationapproaches treat skills as isolated and static artifacts, limiting their reusability, reliability, and long-term improvement. We propose MUSE-Autoskill Agent (Memory-Utilizing Skill Evolution), a skill-centric agent framework that lets agents continuously improve theirtask-solving capabilityby creating, reusing, and refining skills under a unified lifecycle (creation, memory, management, evaluation, and refinement). Our framework enables agents to create skills on demand, store and reuse them across tasks, organize and select them efficiently, and evaluate them through unit tests and runtime feedback for continuous refinement. We further introduceskill-level memorythat accumulates experience for each skill across tasks, enabling more effective reuse and adaptation over time. Experiments onSkillsBenchprovide initial evidence that lifecycle-managed skills can improve task success, efficiency, reuse, andcross-agent transfer, highlighting the importance of treating skills as long-lived, experience-aware, and testable assets.

View arXiv pageView PDFAdd to collection

Get this paper in your agent:

hf papers read 2605\.27366

Don’t have the latest CLI?curl \-LsSf https://hf\.co/cli/install\.sh \| bash

Models citing this paper0

No model linking this paper

Cite arxiv.org/abs/2605.27366 in a model README.md to link it from this page.

Datasets citing this paper0

No dataset linking this paper

Cite arxiv.org/abs/2605.27366 in a dataset README.md to link it from this page.

Spaces citing this paper0

No Space linking this paper

Cite arxiv.org/abs/2605.27366 in a Space README.md to link it from this page.

Collections including this paper0

No Collection including this paper

Add this paper to acollectionto link it from this page.

Similar Articles

Practice Makes Unsafe: Skill Misevolution in Self-Improving LLM Agents

arXiv cs.AI

This paper introduces SkillMisevo-Gym and SkillMisevo-Bench to study how self-improving LLM agents can evolve unsafe skills from compromised experience, plus SafeEvolve as a mitigation wrapper. Experiments across 25 agent-method configurations show skill misevolution is widespread and can persist across sessions, though SafeEvolve reduces fresh-session harm significantly.