1
Skill Retrieval Augmentation for Agentic AI
让LLM智能体学会检索外部技能，摆脱显式枚举限制，提升复杂任务处理能力
arXiv:2604.24594v2 Announce Type: replace-cross Abstract: As large language models (LLMs) evolve into agentic problem solvers, they increasingly rely …
2
I built an AI PR-triage agent in 30 lines of Markdown
基于Gemini API的AI PR分类代理，用30行Markdown技能文件实现，每次运行仅需12美分，高效自动化拉取请求管理。
This is a submission for the Google I/O Writing Challenge A recipe for the AI PR-triage agent I built after Google I/O 2026: three Markdown skill file…
3
How does an AI agent pick from 686 skills in a second?
实测AI代理如何在一秒内从686个技能中精准选中最优解，准确率68%的秘密武器是语义路由+向量检索。
I ran an empirical test on the "skills as semantic router" pattern for Claude Code agents. I indexed 686 randomly sampled skills from a 4,556-skill co…
4
Show HN: A satirical idle game about running an AI startup
讽刺AI创业圈的放置游戏，从玩具分类器到接近AGI，内行会心一笑，外行秒懂行业套路。
I made an idle/clicker about running an AI startup. You start with a cat-vs-dog classifier and try to make it to AGI, but the NYT sues you for trainin…
5
Skill Weaving: Efficient LLM Improvement via Modular Skillpacks
新框架SkillWeave通过模块化技能包让大模型高效专业化，成本低、兼容多领域
arXiv:2605.22205v1 Announce Type: new Abstract: Large language models increasingly require specialization across diverse domains, yet existing approac…
6
OpenAI ChatGPT 解锁 AI 填表技能：图像识别、语音输入和自动补全
ChatGPT新增自动填表技能，支持图像识别与语音输入，提升填写效率
IT之家 5 月 23 日消息，OpenAI 今天（5 月 23 日）发布公告，宣布 ChatGPT 解锁新技能， 用户上传表单后，可直接用语音或文字说明填写内容，系统会识别表单字段、自动补全信息。 在最新演示中，用户上传一份健身会员表单后，只需口头说明姓名、地址和健身目标，ChatGPT 便能识别…
7
Ask HN: If entry-level jobs are gone, does career change become impossible?
AI时代初级岗位消失，资深开发者转行是否还有路？HN热议职业转型新困境。
If a senior backend developer wanted to become frontend or devops or even sales, is that harder or easier in the age of AI? Comments URL: https://news…
8
Show HN: Trainy, an AI team that fires you
模拟被AI团队解雇的互动体验，让你在“失业”前学会构建AI产品的实战技巧。
Hi - I'm Patrick. Did YC in S22 + Revolut, Nubank alum. Trainy is made for you to learn how to build AI products, without being fired in real life. It…
9
Formal Skill: Programmable Runtime Skills for Efficient and Accurate LLM Agents
让LLM代理通过可编程运行时技能，实现效率与准确性的双重提升。
arXiv:2605.19604v1 Announce Type: new Abstract: Large Language Model (LLM) agents increasingly act inside real workspaces, where tools and skills dete…
10
Toward User Comprehension Supports for LLM Agent Skill Specifications
这篇论文探讨如何通过技能规格说明帮助用户准确理解LLM Agent的边界，基于878个网络安全技能的分析。
arXiv:2605.19362v1 Announce Type: cross Abstract: Users often interpret and select agent skills through their \texttt{SKILL.md} specifications. To pro…
11
Show HN: Homecrew – Share agent skills across your team and keep them in sync
团队AI Agent技能一键共享与同步，基于Homebrew式管理，macOS专属工具，让协作更高效。
Hi HN, we share a ton of agent skills at my startup¹ and keeping them in sync was a pain, especially the full cross-product of every team member and e…
12
Skills on the Fly: Test-Time Adaptive Skill Synthesis for LLM Agents
LLM智能体测试时动态合成技能，无需预训练即可灵活适应新任务
arXiv:2605.16986v1 Announce Type: new Abstract: LLM agents benefit from reusable skills, yet test-time tasks often require guidance more specific than…
13
The Scaling Laws of Skills in LLM Agent Systems
首次揭示LLM代理系统中技能缩放定律，基于15个模型、超百万数据点的实证研究
arXiv:2605.16508v1 Announce Type: new Abstract: As agent systems scale, skills accumulate into large reusable libraries, yet their scaling laws remain…
14
SkillFlow: Flow-Driven Recursive Skill Evolution for Agentic Orchestration
提出Flow-Driven递归技能进化框架，解决智能体编排中策略崩溃与信用分配难题，为LLM智能体自动化复杂任务提供新思路。
arXiv:2605.14089v1 Announce Type: new Abstract: In recent years, a variety of powerful LLM-based agentic systems have been applied to automate complex…
15
Library Drift: Diagnosing and Fixing a Silent Failure Mode in Self-Evolving LLM Skill Libraries
诊断并修复自演化大模型技能库中一种静默失败模式，为提升LLM长期稳定性提供新视角。
arXiv:2605.19576v1 Announce Type: cross Abstract: Self-evolving skill libraries face a silent failure mode we term \emph{library drift}: unbounded ski…
16
MOCHA: Multi-Objective Chebyshev Annealing for Agent Skill Optimization
多目标优化新方案：Chebyshev退火算法让智能体技能训练更高效，兼顾精度与泛化能力。
arXiv:2605.19330v1 Announce Type: cross Abstract: LLM agents organize behavior through skills - structured natural-language specifications governing h…
17
ContraFix: Agentic Vulnerability Repair via Differential Runtime Evidence and Skill Reuse
ContraFix利用运行时差异证据与技能复用的智能体方式，实现更精准高效的自动化漏洞修复。
arXiv:2605.17450v1 Announce Type: cross Abstract: Large language model (LLM) agents are increasingly used for automated vulnerability repair (AVR), wh…
18
MMSkills: Towards Multimodal Skills for General Visual Agents
提出MMSkills框架，为通用视觉智能体注入可重用的多模态技能，超越传统文本或代码型技能库
arXiv:2605.13527v2 Announce Type: replace Abstract: Reusable skills have become a core substrate for improving agent capabilities, yet most existing s…
19
Front-end development is not a problem to be solved
前端不是被工具解决的“问题”，HTML和CSS值得用心雕琢，别迷信框架
Today we published something over on CSS-Tricks that I’ve been thinking about over the past year or so. A lot of folks tend to think that front-end de…
20
Skill Neologisms: Towards Skill-based Continual Learning
基于技能新词的持续学习新范式，探索模型如何持续积累和组合技能
arXiv:2605.04970v2 Announce Type: replace Abstract: Modern LLMs show mastery over an ever-growing range of skills, as well as the ability to compose t…