{"slug": "compiling-agent-experience-into-persistent-knowledge-for-skill-evolution", "title": "Compiling Agent Experience into Persistent Knowledge for Skill Evolution", "summary": "Researchers introduced WikiSkill, a framework that co-evolves AI agent skills with a persistent knowledge base (wiki) by separating raw execution experience, accumulated knowledge, and executable skills, and continuously consolidating experience into the wiki. Across diverse benchmarks and models, WikiSkill consistently outperformed state-of-the-art skill-evolution methods and improved over no-skill baselines in most model-benchmark settings, with larger models generally benefiting more from evolved skills and smaller models with skills outperforming substantially larger models without them. The study, submitted to arXiv on 27 Aug 2026, found that evolved skills transfer effectively across models and model families, and that persistent knowledge accumulation in the wiki is critical for effective skill evolution.", "body_md": "# Computer Science > Artificial Intelligence\n\n[Submitted on 27 Aug 2026]\n\n# Title:WikiSkill: Compiling Agent Experience into Persistent Knowledge for Skill Evolution\n\n[View PDF](/pdf/2608.27454)\n\n[HTML (experimental)](https://arxiv.org/html/2608.27454v1)\n\nAbstract:Agent skills package specialized knowledge and workflows into reusable resources that extend AI agent capabilities. Recent work automatically discovers such skills from agent experience, which enables agents to progressively adapt through interaction. However, the insights that guide skill development typically remain scattered across optimization histories, limiting their systematic reuse across iterations. We introduce WikiSkill, a framework that co-evolves agent skills with a persistent knowledge base (wiki). At a high level, WikiSkill separates raw execution experience, accumulated knowledge, and executable skills, while continuously consolidating experience into the wiki, which subsequent skill updates can build on. Across diverse benchmarks and models, WikiSkill consistently outperforms state-of-the-art skill-evolution methods and improves over no-skill baselines in most model-benchmark settings. We find that skill evolution complements model scaling: larger models generally benefit more from evolved skills, while smaller models with skills can outperform substantially larger models without them. We also find that evolved skills transfer effectively across models and model families, and skills evolved by other models can outperform self-evolved skills. Finally, our ablation studies confirm that persistent knowledge accumulation in the wiki is critical for effective skill evolution. These results demonstrate the benefits of systematically accumulating and refining agent experience for developing reusable and transferable skills.\n\n### References & Citations\n\nLoading...\n\n# Bibliographic and Citation Tools\n\nBibliographic Explorer\n\n*(*[What is the Explorer?](https://info.arxiv.org/labs/showcase.html#arxiv-bibliographic-explorer))\nConnected Papers\n\n*(*[What is Connected Papers?](https://www.connectedpapers.com/about))\nLitmaps\n\n*(*[What is Litmaps?](https://www.litmaps.co/))\nscite Smart Citations\n\n*(*[What are Smart Citations?](https://www.scite.ai/))# Code, Data and Media Associated with this Article\n\nalphaXiv\n\n*(*[What is alphaXiv?](https://alphaxiv.org/))\nCatalyzeX Code Finder for Papers\n\n*(*[What is CatalyzeX?](https://www.catalyzex.com))\nDagsHub\n\n*(*[What is DagsHub?](https://dagshub.com/))\nGotit.pub\n\n*(*[What is GotitPub?](http://gotit.pub/faq))\nHugging Face\n\n*(*[What is Huggingface?](https://huggingface.co/huggingface))\nScienceCast\n\n*(*[What is ScienceCast?](https://sciencecast.org/welcome))# Demos\n\n# Recommenders and Search Tools\n\nInfluence Flower\n\n*(*[What are Influence Flowers?](https://influencemap.cmlab.dev/))\nCORE Recommender\n\n*(*[What is CORE?](https://core.ac.uk/services/recommender))# arXivLabs: experimental projects with community collaborators\n\narXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.\n\nBoth individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.\n\nHave an idea for a project that will add value for arXiv's community? [ Learn more about arXivLabs](https://info.arxiv.org/labs/index.html).", "url": "https://wpnews.pro/news/compiling-agent-experience-into-persistent-knowledge-for-skill-evolution", "canonical_source": "https://arxiv.org/abs/2608.27454", "published_at": "2026-08-28 16:18:42+00:00", "updated_at": "2026-08-28 16:48:24.174059+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "ai-agents"], "entities": ["WikiSkill", "arXiv"], "alternates": {"html": "https://wpnews.pro/news/compiling-agent-experience-into-persistent-knowledge-for-skill-evolution", "markdown": "https://wpnews.pro/news/compiling-agent-experience-into-persistent-knowledge-for-skill-evolution.md", "text": "https://wpnews.pro/news/compiling-agent-experience-into-persistent-knowledge-for-skill-evolution.txt", "jsonld": "https://wpnews.pro/news/compiling-agent-experience-into-persistent-knowledge-for-skill-evolution.jsonld"}}