仓库/Skill 库

AGENTS.md 不是写出来的，而是用梯度下降训练出来的You don't write AGENTS.md. You train it with gradient descent.

🦀️ CRAB: 面向多模态语言模型 Agent 的跨环境 Agent 基准。https://crab.camel-ai.org/🦀️ CRAB: Cross-environment Agent Benchmark for Multimodal Language Model Agents. https://crab.camel-ai.org/

Awesome Jev：一份有来源支持的 TypeSafe System One 模型实地指南，包含 SDK、在线 demo、Agent 工具和独立评测。Awesome Jev: a source-backed field guide to TypeSafe's System One model, with SDKs, live demos, agent tools, and independent evaluations.

面向 AI Agent 的上下文工程。token 消耗减少约 80%。解决工具过载问题。内置进程内 BM25 与语义检索的 Skill 与 memory。支持渐进式披露（Progressive Disclosure）。无需向量数据库。Context engineering for AI agents. ~80% fewer tokens. Fix tool overload. Skills and memory with in-process BM25 and semantic retrieval. Progressive Disclosure. No vector DB.

生产级 Go SDK，用于构建具备长期记忆、知识检索与语音能力的 AI agent，可作为库、守护进程或实时管道运行。Production-grade Go SDK for building AI agents with long-term memory, knowledge retrieval, and voice — runnable as a library, a daemon, or a real-time pipeline.

一款 MCP，让 AI 工具安全连接基础设施、编写 IaaS 代码、调试问题并在故障中提供协助，且不危及生产稳定性。专为安全团队审批而设计，为基础设施团队带来如魔法般的体验。An MCP that lets AI tools securely connect to your infrastructure, write IaaS code, debug issues, and assist during incidents - without risking production stability. Built for security teams to approve and infrastructure teams to experience like magic.

端到端人文学科写作助手——一个 Agent Skill（开放的 SKILL.md 格式）。提供 11 种模式，从苏格拉底式研究问题打磨到 AI 使用披露。中英双语（EN/中文），具备学科感知能力（文学/历史/哲学/艺术/宗教/语言学+跨学科）。四层批评、可校准的魔鬼辩护人、声音保留。End-to-end humanities writing assistant — an Agent Skill (open SKILL.md format). 11 modes from Socratic research-question sharpening through AI-use disclosure. Bilingual (EN/中文), discipline-aware (literature/history/philosophy/art/religion/linguistics + cross-disciplinary). Four-layer critique, calibratable devils advocate, voice preservation.

Hermes 桌面端 Bot 模式：一组具名 Agent，各自拥有聊天、头像、任务例程与 Bot 间消息。桌面插件，无需修改核心。Bot Mode for the Hermes desktop: a roster of named agents with their own chats, avatars, routines, and bot-to-bot messaging. Desktop plugin, no core patches.

观察 agent 的思考过程。为 30 种 AI agent runtime 提供零配置可观测性与治理：Claude Code、OpenAI Codex、Hermes、OpenClaw 等 26+。实时 token 成本、会话、tool calls、crons。See your agent think. Zero-config observability & governance for 30 AI agent runtimes: Claude Code, OpenAI Codex, Hermes, OpenClaw & 26 more. Live token costs, sessions, tool calls, crons.

面向你的 AI Agent 的开源控制平面。连接工具、雇佣 Agent、追踪每一个 token 与每一分钱。Open-source control plane for your AI agents. Connect tools, hire agents, track every token and dollar

为 Hermes agents（同时支持 Claude Code 和 Codex）提供由 Jev 驱动的模型路由、记忆、压缩、Skill 选择、计算机与浏览器使用。Jev-powered model routing, memory, compaction, skill selection, computer and browser use for Hermes agents (also Claude Code and Codex)

面向新闻、媒体与学术界的 Claude Code skills，涵盖核查、FOIA、数据新闻与学术写作等场景。Claude Code skills for journalism, media, and academia - verification, FOIA, data journalism, academic writing, and more

Vibe-Coding 容易，DevOps 难。OpenCode 与 Git 为你的应用、业务和个人开发提供 token 高效的 AI agent 自动化。提供面向速度、安全性与 7×24 持续运行的强约定工具、服务、CLI 与 API 栈。开源优先，全面 SOTA。GitHub 协作与团队工作的最佳 AI harness。Vibe-Coding is easy. DevOps is hard. OpenCode & Git token-efficient AI agent automation for your app, business, and personal development. Opinionated tools, services, CLI & API stack for speed, security, and 24/7 results. Open-source first. SOTA everything. The best AI harness for GitHub & teamwork.

一个自托管工作区，让人和编码 Agent 在项目、会话和 Provider 之间协同工作。A self-hosted workspace where people and coding agents work together across projects, sessions, and providers.

面向 AI agent 的最佳浏览器，使用 Rust 构建。Best browser for AI agents, built in Rust.

由 FoJin 驱动的佛教 AI 人格框架——源据可溯、边界清晰、保真可测、开箱即用。FoJin-powered Buddhist AI persona framework — source-grounded, boundary-aware, fidelity-tested, runtime-ready.

面向人类、Agent 和服务的网络A network for humans, agents and services

Pi Coding Agent 扩展的 TypeScript monorepo，涵盖自动化、规划、语言工具、浏览器控制、网络研究、Git 工作流和配置同步。A TypeScript monorepo of Pi Coding Agent extensions for automation, planning, language tooling, browser control, web research, Git workflows, and configuration sync.

安全地为 AI Agent 加速Supercharge AI Agents, Safely

Rome 是 agentic OS。Rome is the agentic OS.

嵌入式 Agent，让客户用于自动化工作、构建视图并连接其工具。Embedded agents your customers use to automate work, build views, and connect their tools.

Jeeves——通过推理改进 Jev 类决策模型Jeeves – Reasoning improves Jev-like decision models

LayerX Network 是一个面向 autonomous agent 的确定性执行与记账网络。LayerX Network is a deterministic execution and accounting network for autonomous agents.

白箱AGI架构探索：元认知（自我认知循环）、持续学习（知识飞轮）、世界模型（条件空间+语义时空图）、自我改进（自举纪律）、零LLM白箱管线与可审计信任护栏。

Secure MCP Tunnel 的客户侧客户端：将私有或 localhost 的 MCP 服务器连接到 ChatGPT、Codex、Responses API 和 AgentKit，无需暴露到公网。Customer-run client for Secure MCP Tunnel: connect private or localhost MCP servers to ChatGPT, Codex, the Responses API, and AgentKit without exposing them to the public internet.

一人即一队 Agent。支持多终端协作的多会话 CLI；/goal 让长任务持续运行；微信/企业微信/飞书网关让你离开后也能随时召回。异步 Python SDK，持久化记忆，自演化 SkillOne person, a team of agents. Multi-session CLI that collaborates across terminals; /goal keeps long tasks running; WeChat/WeCom/Feishu gateway lets you call them back when you walk away. Async Python SDK, persistent memory, self-evolving skills.

为 Hermes Agent 构建语音和文本设备，包含 ESP32 固件、Linux 客户端以及桌面模拟器。Build voice and text devices for your Hermes Agent, with ESP32 firmware, a Linux client, and a desktop simulator.

自托管、兼容 OpenAI 的 AI 网关，用于私有 RAG、自然语言数据访问和工具调用 AgentSelf-hosted, OpenAI-compatible AI gateway for private RAG, natural-language data access, and tool-calling agents.

面向医学研究的 Agent Skills——文献检索、报告规范与引用核查、统计、发表图表、投稿。兼容 Claude Code、Codex、Cursor 与 GitHub Copilot。由医师-研究者构建，在真实发表论文上验证。MIT 协议。Agent Skills for medical research — literature search, reporting-guideline & citation checks, statistics, publication figures, submission. Works with Claude Code, Codex, Cursor & GitHub Copilot. Built by a physician-researcher, tested on real publications. MIT.

2007 年的诺基亚已无法搜索 Google，于是作者为其接入了 Claude、ChatGPT、Gemini 与 Grok。面向 Nokia S40/S60 的 J2ME 应用 + 轻量 Go 服务器。A 2007 Nokia can't search Google anymore, so I gave it Claude, ChatGPT, Gemini and Grok. J2ME app for Nokia S40/S60 + tiny Go server.

ResearchClaw 是面向研究的个人 AI 助手：部署便捷，支持本地或云端运行，可与常用聊天应用集成；通过可扩展 skills，端到端简化文献综述、笔记、实验追踪与论文写作ResearchClaw is a personal AI assistant built for research: fast to set up, easy to run locally or in the cloud, and ready to integrate with the chat apps you already use. With extensible skills, it helps you streamline literature review, note-taking, experiment tracking, and paper writing—end to end.

AIにまじなPPTXを作らせるClaude Codeスキル — スライド規約＋62型スライド型カタログ（SlideSpec 36型＋自由記述27パーツ）＋生成パイプライン＋機械チェック

agent 协调的开放标准The open standard for agent coordination

Adobe Premiere Pro MCP：独立、本地优先的服务端，支持对应工作流。通过本地 CEP 桥接连接 Claude、Codex 或 Cursor。Adobe Premiere Pro MCP — independent, local-first server for supported workflows. Connect Claude, Codex, or Cursor through a local CEP bridge.

面向研究工作流的通用 SKILL.md 目录：覆盖文献综述、研究设计、项目记忆、稿件写作与跨 Agent 委托，兼容 Claude Code、Codex、Gemini、Cursor、OpenClaw 及通用 AI 客户端Universal SKILL.md catalog for research workflows: literature review, research design, project memory, manuscript writing, and cross-agent delegation for Claude Code, Codex, Gemini, Cursor, OpenClaw, and generic AI clients.