SkillOps: Managing LLM Agent Skill Libraries as Self-Maintaining Software Ecosystems
cs.SE, cs.MA
Submitted: 2026-05-13
Updated: 2026-11-07
Comments: 23 pages, 9 figures. Submitted to NeurIPS 2026
Code: https://github.com/Hik289/SkillOps
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- CUA-Skill: Develop Skills for Computer Using Agent
- LEGOMem: Modular Procedural Memory for Multi-agent LLM Systems for Workflow Automation
- Graph-of-Skills: Dependency-Aware Structural Retrieval for Massive Agent Skills
- Test-Driven Development for Code Generation
- VeriGuard: Enhancing LLM Agent Safety via Verified Code Generation
- GPT-4o System Card
- ToolLLM: Facilitating Large Language Models to Master 16000+ Real-world APIs
- SKILLFOUNDRY: Building Self-Evolving Agent Skill Libraries from Heterogeneous Scientific Resources
- HuggingGPT: Solving AI Tasks with ChatGPT and its Friends in Hugging Face
- ALFRED: A Benchmark for Interpreting Grounded Instructions for Everyday Tasks
- ALFWorld: Aligning Text and Embodied Environments for Interactive Learning
- SkillX: Automatically Constructing Skill Knowledge Bases for Agents
- Reinforcement Learning for Self-Improving Agent with Skill Library
- ReWOO: Decoupling Reasoning from Observations for Efficient Augmented Language Models
- A-MEM: Agentic Memory for LLM Agents
- ReAct: Synergizing Reasoning and Acting in Language Models
Related papers
- ArkEval: Benchmarking and Evaluating Automated CodeRepair for ArkTS
- TestPrism: Rethinking Test Evaluation Beyond a Single Reference
- Mining Workflow Graphs for Black-Box Boundary Testing of Conversational LLM Agents
- Falsification-Based Verification of LLM-Generated Optimization Models: Sound Test Batteries and Their Detection Limits
- GitSkills: A Dataset of Agent Skills on GitHub
- For Coding Agents, the Repository Matters