热门好用的 Agent Skill 推荐
用于审查、研究、写作和数据分析的可复用智能体工作流。
一组面向中文内容运营场景的 agent skills，覆盖微信公众号、小红书、抖音、视频号和 X/Twitter 的合规起号、定位、选题、内容计划和复盘。
A compact agent skill for turning short prompts into beautiful, realistic GPT Image 2 images.
52 AI-powered skills across 8 stages with a closed-loop flywheel. Research programs, scout trending content, write data-backed posts, generate infographics, build pages, deploy, track, optimize,...
Install and use the measurable adversarial code-audit skill with scan-only defaults and explicit permission before source edits.
Swift Testing Agent Skill for AI Coding Assistants
Drag-and-drop email editor for modern apps — source-available, MIT after two years
新论文层出不穷，英文读得再顺也不如母语省心；我们想把这些论文变成排版规范、读着顺口的中文 PDF。
一个适用于 Claude Code / Codex 的技能，用来生成惊艳、带动画、单文件的 HTML 演示文稿，并内置浏览器编辑器：可拖拽对象、调整大小、编辑文本、重排页面、本地保存，并导出干净的独立 HTML。
The script. The prompt. The hosts' roles. Their voices. The content they discuss. All of it.
iDeer 是一个**工具。你告诉它你关注什么，它替你盯住散落在各个平台的更新，用 LLM 做筛选、打分、摘要，最后把值得看的内容在你设定的时间自动送到邮箱里。
Direct complete videos—from the first idea to production-ready media.
Beautiful README badges, inspired by shields.io and styled like shadcn/ui.
AI video skill for Claude Code & Codex — cinematic product videos with Remotion: 152 shot recipe cards, 209 motion previews, a production-ready template
11 skills that help Claude Code and Codex write LinkedIn posts, comments, and replies in your voice. They draft content, strip AI tells, and wait for your approval before anything gets published.
bggg-skills 是 BGGG 开源的 Codex Skills 集合仓库。
Everything distilled from — a public X trader and AI/semiconductor supply-chain analyst known for tracing hyperscaler capex into upstream bottlenecks.
科研绘图skill、论文绘图skill、图片转visio等可编辑格式，将生成图转化为论文可编辑图，便于作者调整绘图细节
An open-source Dynamic Workflow runtime for building observable, reusable Agent graphs.
Community-driven AI agent skills for Claude Code, OpenCode, Gemini CLI, Cursor, VS Code Copilot, Codex, Windsurf, Antigravity, and other AI assistants.
SkVM is a compilation and runtime system that makes LLM agent skills portable across heterogeneous models and harnesses. It has four major parts:
A community-driven collection of DFIR / incident response : reusable prompts, workflows, and helper files that help practitioners move faster, stay consistent, and maybe—just maybe—get some sleep.
Built by a physician-researcher, tested on real publications.
Skills distilled from the Lab's real work and collaboration practices
A self-evolving WorldQuant BRAIN alpha research skill. It helps agents design WQ Alpha expressions, search fields, diagnose simulation failures, check IS metrics, manage self-correlation, and turn...