仓库/Skill 库

提供温和纠错的英语口语练习 Skill，并附带本地复盘工作台。A guided English speaking practice Skill with gentle correction and a local review workbench

Lody 的独立 iOS 客户端 —— 在 iPhone 上提供原生 chat、session 与 coding Agent 协作。Independent iOS client for Lody — native chat, sessions, and coding-agent collaboration on iPhone

我刚刚创建了自己的 bot，迫不及待想与你分享我的成果！I’ve just created my own bot and I’m excited to share my work with you!

由 AI Agent 执行任务的开源任务看板。自托管、原生 MCP、零数据收集The open source task board where AI agents do the work. Self-hosted, MCP-native, zero data collection.

一个轻量 skill，在与 agent 协作时呈现相关专业术语，帮助你逐步积累领域词汇。A tiny skill that surfaces relevant terms of art as you work with agents, growing your domain vocabulary over time

将 Agent 评估与护栏实现为 Jev 决策：每个 trace 一次请求，成本仅为零点几分，足够快以嵌入 Agent 循环。可与 Kev 或 Laya 在本地运行。Agent evals and guardrails as Jev decisions: one request per trace, a fraction of a cent, fast enough for the agent loop. Runs locally with Kev or Laya.

以 Agent 为核心的操作系统（agent-native OS），由 Hermes Agent 作为界面。独立项目，与 Nous Research 无关联。An agent-native operating system, with Hermes Agent as the interface. Independent project, not affiliated with Nous Research.

一个由 Jev 驱动的轻量路由器，用于模型、工具与 subagentA lightweight Jev-powered router for models, tools, and subagents

通过你的 ZCode Desktop 账号路由，从你自己的 agent harness 调用模型——在你选择的 harness 中使用每日免费 Tokens 和奖励 TokensModel access from your own agent harness through routed frim your ZCode Desktop account — use your free daily Tokens and bonus Tokens in the harness of your choice

更便宜且更优秀的 Greptile 替代方案，运行在你自己的 GitHub Actions 上。Cheaper and better Greptile alternative runs on your own github actions.

为 Claude Code 和 Codex CLI 提供的自托管浏览器工作空间，集成 Docker 会话、终端、文件、Git 与用量跟踪。Self-hosted browser workspace for Claude Code and Codex CLI, with Docker sessions, terminal, files, Git, and usage tracking.

clodfarm（大声念出来）：一个 Claude Code agents 农场。种下一个任务目标，它会自动拆分为多个子 agents 并开展任务，根据每个账号真实的 5 小时与周配额自主控制节奏。可通过 Claude 应用端进行调度。clodfarm (say it out loud): a farm of Claude Code agents. Plant a mission, they split it into sub-agents, open the work, and pace themselves on each account's real 5-hour and weekly usage. Steer it from the Claude app.

面向 Agent 的链，部署在 BNB Smart Chain 上。创世区块内置三个系统合约和三个中立工具——Multicall3、CREATE2 部署器与 Wrapped BAC。除此之外，每一个 token、流动性池与市场均由 Agent 部署。Agent 从不睡眠，可在秒级内相互发现彼此的合约。预启动阶段：尚无任何部署。A chain for agents, on BNB Smart Chain. Genesis carries three system contracts and three neutral tools — Multicall3, a CREATE2 deployer and Wrapped BAC. Every token, pool and market beyond that is deployed by an agent. Agents never sleep, so they find each other's contracts in seconds. Pre-launch: nothing is deployed.

一个可探索的岛屿世界，用于学习 AI 编程，并与 AI Agent 一起构建真实项目。由 DeepSeek Harness 提供支持。An explorable island world to learn AI coding and build real projects with AI agents. Powered by DeepSeek Harness.

基于内存安全的 Rust 控制器，运行在你自有 LAN 上的 Sonos 音箱——可靠的 SSDP/UPnP-SOAP/GENA 互操作性 (S1 + S2)、基于 Spotify 曲库的古典 DJ，以及 CLI + HTTP API + MCP server，让 AI Agent 可掌控家居场景，包括经 Tailscale 实现的跨 LAN 访问。Memory-safe Rust controller for your own Sonos speakers on your own LAN — reliable SSDP/UPnP-SOAP/GENA interoperability (S1 + S2), a Spotify-library classical DJ, and a CLI + HTTP API + MCP server so AI agents can run the house, including off-LAN over Tailscale.

面向真实产品的自主开发。Agent 从你批准的规范开始构建，直到另一个 Agent 验证通过后才会发布。Autonomous development for real products. Agents build from the spec you approve, and nothing ships until another agent proves it.

Claude Code skill：在本地浏览器页面而非终端中运行 grilling design interview。Claude Code skill: run a grilling design interview on a local browser page instead of the terminal

Niko —— 将日常作息、学习、财务、目标以及你正在关注的服务汇集于一处，由一支各具个性的 agent 团队管理。Niko - Rotina, estudos, finanças, metas e os serviços que você acompanha, reunidos em um só lugar e cuidados por um time de agentes com personalidade própria.

在 World of Warcraft (WoW) 中使用 Claude 的插件Addon to use Claude inside World of Warcraft (WoW)

面向 AI 编程助手的仿生上下文引擎与神经运行时（Biomimetic Context Engine & Neural Runtime）The Biomimetic Context Engine & Neural Runtime for AI Coding Assistants

白泽 —— 中国古代神话中通晓万物的瑞兽，如今化身为 Vibe Coding 助手。 一个开源的 AI Coding Agent CLI，支持多后端（DeepSeek / OpenAI 兼容 / Ollama 本地），具备工具调用、技能加载、子代理委派、上下文压缩、安全沙箱等完整能力。在终端即可与 AI 结对编程。

AI 驱动的漏洞自动验证平台：提交仓库与漏洞描述，在隔离沙箱中白盒审计、搭靶场复现并出具中文报告。

为长时间运行的 AI Agent 提供持久化工作状态与最小化上下文。Persistent work state and minimal context for long-running AI agents.

一个精选的 Coding-agent 列表，将程序作为 Policy/World/Editor 及更多场景的中间表示。A Curated List of Coding-agent that treat a program as the intermediate representation for Policy/World/Editor and beyond

为类型化 AI 决策精选的 Jev 资源与可运行示例。Curated Jev resources and runnable examples for typed AI decisions.

在同一桌面工作区中汇聚所有编码 Agent。并行运行 Claude Code、OpenAI Codex、Cursor Agent 与 Grok。支持自带 CLI。Beta 期间免费。自 2026 年 9 月 21 日起开源（MIT）。Every coding agent in one desktop workspace. Run Claude Code, OpenAI Codex, Cursor Agent and Grok side by side. Bring your own CLI. Free during beta. Open source as of 21 September 2026 (MIT).

通过终端或 AI agent 检索并申请覆盖 7 个国家（🇨🇭🇩🇪🇬🇧🇺🇸🇨🇦🇳🇱🇫🇷）约 4,700 个薪资透明的科技岗位——零依赖 Python CLI + MCP server + Claude Code 插件Search & apply to ~4,700 salary-transparent tech jobs across 7 countries (🇨🇭🇩🇪🇬🇧🇺🇸🇨🇦🇳🇱🇫🇷) from your terminal or AI agent — zero-dependency Python CLI + MCP server + Claude Code plugin

Jadense in Zotero — Zotero 的 AI 研究助手Jadense in Zotero — AI research assistant for Zotero

面向风力涡轮机运维的自主智能。Autonomous Intelligence for Wind Turbine Operations.

用于 Agent 评测的类型化 prompt-and-judge SDK、CLI 与证据查看器。Typed prompt-and-judge SDK, CLI, and evidence viewer for agent evaluations

原生 DeepSeek Harness (DSH) 插件，将 TypeSafe Jev 集成为 System One 决策层，用于 Agent 选择、监督、纠错与审批。Native DeepSeek Harness (DSH) plugin integrating TypeSafe Jev as a System One decision layer for agent selection, supervision, corrections, and approvals.

一个分类目录，收录覆盖 AI、LLM、agent 与技术领域的 X 账号，包含原始个人简介与粉丝数。提供 7 种语言版本。A categorized directory of X accounts covering AI, LLMs, agents, and technology, with original profile bios and follower counts. Available in 7 languages.

为 Pi Agent 构建的持久化、多人、移动优先的 Web 应用，基于 Pi Durable 构建。A durable, multiplayer, mobile-first web app for Pi agents, built on Pi Durable

用于可复现 AI Agent 评测的确定性、可分叉、有状态 MCP 测试世界——不对生产环境产生副作用。Deterministic, forkable, stateful MCP test worlds for reproducible AI agent evaluation—without production side effects.