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。

5433个结果
5433 个结果
TZ
t-zevin/skillguardrail
开发工具

SkillGuardrail is an open-source pre-install security scanner and guarded installer for Agent Skills used by Codex, Claude Code, Cursor, Gemini CLI, and OpenClaw.

14570
OS
obie/skills
开发工具

Production-ready skills for enhanced Claude Code development workflows.

9570
SS
shelby/shelby-skills
开发工具

Agent skills for building with the Shelby ecosystem. Each skill provides structured documentation that AI coding agents (Claude Code, Cursor, Copilot, etc.

4470
CR
conway-research/skills
开发工具

Community-maintained skills for AI agents to help them save tokens and write to the world.

4270
PA
paulbratslavsky/astro-strapi-example-project
开发工具

This is an example project for Astro 6 and Strapi 5 (latest). It is a simple project that uses Astro and Strapi to create a website with a landing page, a blog, and a contact form.

10270
LG
leon-gittech/skill2poster
浏览器自动化

Codex skills for academic poster generation: turn paper content into accurate A0 HTML/PDF posters via AI design mockups.

7070
MS
modular/skills
开发工具

These are the official AI agent skills from Modular for working with the Modular Platform, including MAX and Mojo. They follow the Agent Skills Standard.

14870
MC
microsoft/cat-agent-skills
开发工具

A community gallery of — reusable instruction sets (and optional script bundles) you can drop into , , and agents. The site is inspired by pcf.

3870
OP
opensearch-project/opensearch-agent-skills
开发工具

A collection of Agent Skills for building search applications and analyzing observability data with OpenSearch.

3770
ZL
zeroclaw-labs/zeroclaw-skills
开发工具

A skill registry for ZeroClaw — AI agent skills, tools, and workflows. Skills follow the agentskills.io open specification.

4970
RR
runxhq/runx
开发工具

A skill is expertise published as a portable SKILL.md: an operating manual that a human can understand and an agent can act from. Skills compose into graphs and real work without bespoke glue code.

7570
SS
stellar/stellar-dev-skill
开发工具

A comprehensive AI skill for modern Stellar development with current best practices. https://skills.stellar.org/

4970
GG
growthack88/growth-marketing-os
开发工具

— the first bilingual AI marketing asset library on GitHub.

8670
KB
kianzzz/book-sales-video
开发工具

一个面向 Codex/Agents 的中文 Skill:从书名或飞书多维表格中的成稿文案出发,完成书籍核验、语义分镜、豆包配音、Pexels 开场、AI 配图、双语字幕和 OpenChatCut 可编辑视频初稿。

14070
US
unitoneai/securityskills
开发工具

AI coding agents can perform security reviews, but they hallucinate framework control numbers, miss entire vulnerability categories, and produce inconsistent output across runs.

3470
DT
dotanminh/travel-optimization-engine
开发工具

Bộ 8 AI Skills tự động hóa toàn bộ quy trình tối ưu chi phí vé máy bay. Tiết kiệm 30-66% cho cá nhân/gia đình, 10-15% cho doanh nghiệp.

3870
FC
forefy/.context
开发工具

An optimized collection of AI agent skills, loops and workflows for security auditing

12870
PA
perplexityai/api-platform-developers
开发工具

Official Perplexity API Platform skills, published as portable Agent Skills. They work in Claude Code, OpenAI Codex CLI, and any other Agent Skills client, and the repository doubles as a Claude Code...

8270
YD
yc-duan/api-image
API 开发

它的作用很简单:当你在 API 形式使用 Codex App,没法直接用官方内置 imagegen 时,让 Codex 改走你自己配置的 OpenAI-compatible 图片接口,例如通过 CPA 反代出来的 gpt-image-2。

8970
RO
render-oss/skills
开发工具

Render Agent Skills

8070
OF
omerbsezer/fast-llm-agent-mcp
开发工具

This repo covers LLM, Agents, MCP Tools, Skills concepts both theoretically and practically: - LLM Architectures, RAG, Fine Tuning, Agents, Tools, MCP, Agent Frameworks, Reference Documents.

8770
AM
alchaincyf/mrbeast-skill
开发工具

"I don't think of myself as a YouTuber. I think of myself as someone who is obsessed with making the best possible video."

10070
KD
kangarooking/design-image-studio
API 开发

design-image-studio 是一个面向创意与设计场景的 Agent Skill。它不是简单地把一句话塞给文生图模型,而是先读取并保留完整的 Claude 设计系统提示词,再把需求整理成设计推理和结构化 brief,最后把这些设计判断压缩成更强的生图 Prompt,并直接调用 Volcengine Seedream 生成图片。

9770
DC
dukesun99/corpus2skill
研究分析

This is the official implementation of the paper (Sun, Wei, and Hsieh, 2026), accepted to .

8270