共 118 条资源，覆盖 Agent 框架、Skill 技能包、MCP 工具与提示词教程。支持搜索、排序与收藏。
全球用户量最大的 AI 对话助手：写作、翻译、总结、头脑风暴、数据分析全能，免费版够用，Plus 解锁 GPT-4o 与文件上传，是 AIGC 入门的第一站。The world's most popular AI assistant for writing, translation, summarization and brainstorming — the first stop for AIGC beginners.
艺术质感最强的文生图工具：一句话生成海报级画面，V7 版本对中国风、产品摄影、插画都有极强表现，设计师与电商人的主力工具。The most artistic text-to-image tool: poster-grade visuals from one prompt, strong at product shots, illustration and Chinese aesthetics.
面向中文读者的深度学习交互式教材：概念、数学、可运行代码三位一体，从线性回归一路讲到注意力机制与优化算法，配套 B 站视频课程，被国内高校广泛采用。An interactive deep learning book for Chinese readers: concepts, math and runnable code in one place, from linear regression to attention, with companion video courses.
斯坦福吴恩达的机器学习旗舰课程新版本：用 Python 与 scikit-learn 教学，覆盖监督学习、聚类、异常检测、推荐系统与决策树，配套 Labs 与测验，是全球公认的 ML 第一课。Andrew Ng's updated Machine Learning Specialization: Python-based, covering supervised learning, clustering, anomaly detection and recommender systems with hands-on labs.
本地大模型一键运行器：一条命令拉取并运行 Llama/Qwen/DeepSeek 等开源模型，自动匹配 GPU，提供 OpenAI 兼容 API，是本地化部署事实标准。Run open-source LLMs locally with one command: Llama, Qwen, DeepSeek and more, with GPU auto-detection and an OpenAI-compatible API.
前 Tesla AI 总监 Andrej Karpathy 的「从零到一」系列：从反向传播手推开始，逐步用 PyTorch 手写 micrograd、makemore、GPT，最终从零实现一个类 GPT-2 的语言模型——理解 Transformer 内部原理的最佳路径。Andrej Karpathy's Zero-to-Hero series: from hand-deriving backprop to building GPT-2 from scratch in PyTorch — the best path to truly understand Transformers.
以文风自然著称的 AI 助手：长上下文能一次吃下整本书，中文写作不像翻译腔，Artifacts 直接生成可预览的文档和网页，写作质量派首选。Anthropic's assistant known for natural writing style, huge context window and Artifacts live previews — the quality pick for writers.
OpenAI 官方轻量 Agent 框架：以 Agent、Handoff（任务移交）、Guardrail（安全护栏）三个原语构建多 Agent 系统，代码量少、心智负担低，是 Swarm 的生产级继任者。OpenAI's lightweight agent framework built on three primitives — agents, handoffs and guardrails — the production-ready successor to Swarm.
GitHub 40k+ star 的大模型学习路线图：按 Fundamentals → LLM Scientist → LLM Engineer 三阶段组织，每步附论文、Colab 与资源清单，从原理到量化部署全链路覆盖。A 40k+ star LLM learning roadmap organized in three tiers: Fundamentals, LLM Scientist and LLM Engineer, with papers and Colabs for every step.
国产 AI 视频头部选手：图生视频动态自然、人物动作连贯，支持首尾帧控制和运镜指令，免费额度大方，短视频创作者首选。A leading Chinese AI video generator with natural motion, first/last frame control and camera movement prompts — generous free tier.
GitHub 官方 MCP Server：让 Agent 直接读写仓库、管理 Issue/PR、检索代码与 Actions 状态，是把 AI 接入研发流程的第一步。GitHub's official MCP server: read/write repos, manage issues and PRs, search code and inspect Actions from your agent.
Hugging Face 官方学习中心：Transformers、Diffusion、RLHF、Agents 等多门免费课程，全部配套在线可运行环境，是上手 transformer 生态事实标准工具链的官方路径。The official Hugging Face learning hub: free courses on Transformers, Diffusion, RLHF and Agents with runnable notebooks.
Datawhale 出品的中文 LLM 应用开发教程：从 API 调用到 Prompt 工程、RAG 知识库、Vector 检索、LangChain 应用与部署，全程围绕一个「知识库助手」项目实战。Datawhale's Chinese LLM application course: from API calls to prompt engineering, RAG, vector search and LangChain, built around one knowledge-base assistant project.
一句话生成完整歌曲（含人声演唱）：写词作曲编曲混音全包，中文歌效果惊艳，做视频 BGM、品牌歌、生日祝福歌的宝藏工具。Turn one sentence into a full song with vocals — lyrics, composition and mixing included. Great for BGM and custom songs.
以数据为中心的 LLM 框架：RAG 索引、检索、文档解析能力业界最全，现已扩展出 Agent 工作流（Workflows），是知识库类应用的首选底座。The data-centric LLM framework: the richest RAG indexing/retrieval toolchain, now extended with agent workflows.
Jeremy Howard 的实战深度学习课程：第一节课就训练出可用模型，再逐步深入原理。理念是「先跑起来再理解」，配套 fastai 库把最佳实践封装到几行代码。Jeremy Howard's top-down deep learning course: train a working model in lesson one, then dig into how it works. Comes with the fastai library.
字节出品的中文生图/生视频平台：直接用中文描述出图，懂"国潮""水墨"这些文化词，与剪映生态打通，国内用户最顺手的创作入口。ByteDance's Chinese-native image and video generation platform that understands cultural terms, integrated with CapCut ecosystem.
PyTorch 官方教程中心：从 60 分钟闪电入门到训练技巧、部署、量化，全链路示例代码可直接在 Colab 运行，学框架的第一手资料。The official PyTorch tutorial hub: from the 60-minute blitz to training recipes, deployment and quantization, all runnable in Colab.
国民剪辑软件的 AI 全家桶：智能字幕、AI 配音、图文成片、一键剪同款，视频创作者从素材到成片的一站式流水线。CapCut's AI suite: auto captions, AI voiceover, text-to-video and templates — an end-to-end pipeline for creators.
月之暗面出品的国产助手：超长中文文本处理是招牌，联网搜索+文档解析开箱即用，中文场景理解到位，免费且访问无门槛。Moonshot's Chinese assistant excelling at long-document Q&A and web search, free and reliable in China.
Pydantic 团队出品的 Agent 框架：用类型注解约束模型输出，把「结构化输出 + 依赖注入 + 可测试」做到极致，工程化程度高，适合严谨的后端项目。Agent framework from the Pydantic team: typed, validated outputs with dependency injection and testability — built for serious backend engineering.
斯坦福研究生级机器学习课程：严格数学推导 + 精炼 notes，覆盖 GLM、SVM、学习理论、EM、强化学习，吴恩达主讲版本是经典中的经典，适合打深基础。Stanford's graduate-level ML course with rigorous math derivations and concise notes, covering GLMs, SVMs, learning theory, EM and RL.
斯坦福 NLP 旗舰课：从词向量到 Transformer、预训练与提示，Manning 主讲，几乎是大模型时代的「前置必修课」，讲义与作业质量双高。Stanford's NLP flagship course from word vectors to Transformers and prompting, taught by Christopher Manning — the prerequisite course for the LLM era.
Google 官方开源的 Gemini 命令行智能体：把 Gemini 模型带进终端，支持工具调用、MCP 连接与多模态问答，是继 Claude Code 之后编程 Agent 领域的官方级选手。Google's official open-source Gemini CLI agent: bring Gemini models into your terminal with tool use, MCP support and multimodal Q&A.
字节跳动的全民 AI 助手：写作、绘图、语音对话、智能体广场一体，移动端体验极佳，完全免费，适合日常轻度使用和内容创作起步。ByteDance's all-in-one free AI assistant with writing, image generation and voice chat — great for mobile users.
终端里的结对编程 Agent：直接修改你本地仓库代码并自动生成规范的 Git 提交，支持代码库地图与多种模型，是 CLI 编程 Agent 的开山之作。Pair-programming agent in your terminal: edits your local repo and commits cleanly, with repo mapping and multi-model support.
斯坦福王牌 CV 课程：从图像分类与 kNN 讲到 CNN、检测、分割、3D 与生成模型，课程 notes（中文翻译版流传极广）质量极高，Fei-Fei Li 团队出品。Stanford's flagship computer vision course: from image classification to CNNs, detection, segmentation and generative models, with legendary course notes.
Datawhale 的开源大模型部署手册：环境配置、Llama/Qwen/ChatGLM 等主流开源模型的部署、推理、LoRA 微调全流程，中文环境踩坑指南性质极强。A Chinese hands-on guide to deploying open-source LLMs: environment setup, inference and LoRA fine-tuning for Llama, Qwen, ChatGLM and more.
专业级 AI 视频工具箱：文/图生视频、视频修补、运动笔刷、绿幕抠像一应俱全，影视广告行业实际在用的生产力。The professional AI video suite: Gen-4 generation, inpainting, motion brush and green screen — widely used in film and ad production.
最流行的 LLM 应用开发框架，提供链式调用、Agent 编排、工具调用与 RAG 全套能力，Python 与 JS 双语言生态，适合快速搭建各类 AI Agent 应用。The most popular framework for building LLM applications: chains, agent orchestration, tool calling and RAG out of the box, in both Python and JavaScript.
基于图（状态机）的 Agent 编排框架，擅长多 Agent 协作、循环与人工介入流程，是 LangChain 官方推荐的复杂 Agent 生产方案。Graph-based agent orchestration from the LangChain team. Ideal for multi-agent collaboration, loops and human-in-the-loop workflows in production.
开源 LLM 应用开发平台：可视化编排工作流、内置 RAG 知识库、Agent 与模型管理，国产开源明星项目，企业落地案例多。Open-source LLM app platform with visual workflow, built-in RAG knowledge base, agents and model management.
Anthropic 官方 Agent Skills 仓库，包含 docx/pptx/xlsx/pdf 等文档处理技能及示例，是学习 Skill 规范（SKILL.md）写法的最佳范本。Anthropic's official Agent Skills repository with document skills (docx/pptx/xlsx/pdf) and examples — the reference for SKILL.md conventions.
Anthropic 推出的模型上下文开放协议，已成为 AI 工具调用的事实标准；官方 servers 仓库收录文件系统、搜索、数据库等参考实现。Anthropic's open protocol connecting AI models to tools and data — the de-facto standard, with reference server implementations.
Anthropic 官方终端编程 Agent：理解整个代码库、跨文件修改、执行命令与 Git 操作，支持 MCP 与自定义 Skills 扩展。Anthropic's official terminal coding agent: codebase-aware edits, shell commands and git workflows, extensible via MCP and Skills.
最系统的 Prompt 工程免费教程（有中文版）：从零样本到 CoT、ReAct、Agent 应用，配套论文清单，适合系统学习。The most systematic free prompt engineering guide (Chinese version available): zero-shot to CoT, ReAct and agent patterns.
微软官方 AI Agent 入门课程（11 课，中文讲义）：从工具使用、RAG、规划到多 Agent 设计模式，带代码作业。Microsoft's official beginner course (11 lessons, Chinese notes) covering tool use, RAG, planning and multi-agent design patterns.
Google 开源的 Agent Development Kit：原生对接 Gemini 与 Vertex AI，内置多 Agent 层级编排、工具生态与评测能力，适合构建企业级 Agent 服务。Google's open-source Agent Development Kit with native Gemini/Vertex AI integration, hierarchical multi-agent orchestration and built-in evaluation.
开源 LLM 工程可观测平台：链路追踪、成本统计、Prompt 版本管理、数据集与自动评测，Agent 上线后必备的「仪表盘」。Open-source LLM observability: tracing, cost tracking, prompt versioning, datasets and automated evals.
周志华《机器学习》：国内 ML 教材的事实标准，覆盖十余种经典算法，语言凝练、体系完整；配合《南瓜书》的公式推导更佳，是考研与面试的高频出处。Zhou Zhihua's Machine Learning (the 'Watermelon Book'): the de-facto standard ML textbook in China, covering classic algorithms with rigor.
高吞吐 LLM 推理引擎：PagedAttention + 连续批处理大幅提升并发吞吐，OpenAI 兼容 API，生产环境自托管模型服务的首选。A high-throughput LLM inference engine with PagedAttention and continuous batching; OpenAI-compatible API, the go-to for self-hosted serving.
藏在 ChatGPT 里的生图：中文描述直接出图，最大的优势是"对话式改图"——说一句"把背景换成雪天"它就改，零学习成本。Image generation inside ChatGPT: describe in natural language and refine by chatting — zero learning curve.
给 Agent 喂「最新版本文档」的 MCP 服务：按库名实时拉取对应版本的 API 文档，解决模型知识过时导致的错误调用，写代码必备。An MCP service that feeds agents up-to-date, version-specific library docs — fixes stale-knowledge API mistakes.
吴恩达与 Isabel Kloumann 合作的提示词工程短课：总结、推断、转换、扩展四大范式，配套 Jupyter 实操，两小时建立提示词的工程化思维。A short prompt engineering course by Andrew Ng: four paradigms — summarize, infer, transform, expand — with hands-on Jupyter notebooks.
开源节点式生图工作流引擎：把模型、采样、控制条件连成流水线，可控性和复现性最强，本地跑免费用，进阶玩家的终极选择。Open-source node-based image generation workflow engine with maximum control and reproducibility — the power user's choice.
OpenAI 官方终端编程 Agent：本地沙箱执行、支持多模型与审批策略，与 Claude Code 同台竞争的 CLI 选手。OpenAI's official terminal coding agent with sandboxed execution, multi-model support and approval policies.
中文世界最系统的免费提示词教程：从基础结构、进阶技巧到 ChatGPT/ Midjourney 实战，配套可运行示例，适合零基础系统入门。The most systematic free Chinese prompt-engineering tutorial, from basics to practical ChatGPT and Midjourney workflows.
李航《统计学习方法（第二版）》：以统计视角严格推导感知机、SVM、EM、CRF 等经典模型，篇幅精炼、推导完整，与西瓜书并称国内 ML 双璧。Li Hang's Statistical Learning Methods: rigorous derivations of perceptron, SVM, EM and CRF from a statistical perspective; concise and complete.
统一 LLM 网关：把 100+ 家模型 provider 统一成 OpenAI 格式，带配额、预算、限流与 fallback 路由，生产环境管理多模型调用的标配。A unified LLM gateway: 100+ providers normalized to the OpenAI format, with quotas, budgets, rate limits and fallback routing.
国产开源 RAG 引擎：以深度文档解析见长（表格、扫描件、复杂版式），提供可视化知识库与可溯源引用，中英文场景表现好。Open-source RAG engine with deep document parsing (tables, scans, complex layouts) and traceable citations.
模型聚合路由：一个 API key 访问主流所有商业与开源模型，统一计费、支持比价与自动降级，免管理多家账号的烦恼。One API key for all major models: unified billing, price comparison and automatic fallback across commercial and open models.
没有广告的 AI 搜索：直接给出结构化答案并标注来源，学术模式可搜论文，写报告查资料的效率是传统搜索的数倍。An AI search engine that directly generates structured answers with citations — a research accelerator for reports and papers.
开源的 IDE AI 助手（VS Code / JetBrains）：可接任意模型（含本地 Ollama），支持自定义规则、上下文提供者与 Agent 模式，替代闭源 Copilot 的首选。Open-source IDE assistant for VS Code and JetBrains: any model (including local Ollama), custom rules and agent mode.
HuggingFace 出品的极简 Agent 库：主打 CodeAgent——让模型直接写 Python 代码作为行动（比 JSON 工具调用更灵活），几行代码即可搭建带工具的 Agent。HuggingFace's minimalist agent library featuring CodeAgents that write Python code as actions, more flexible than JSON tool calls.
OpenAI 官方提示词工程指南：策略与案例覆盖指令清晰度、分隔符、少样本、思维链与系统化测试，是写 GPT 提示词的权威参考。OpenAI's official prompt engineering guide: strategies and examples covering clarity, delimiters, few-shot, chain-of-thought and systematic testing.
以画面质感著称的 AI 视频工具：物理光影真实、镜头运动电影感强，Ray2 模型对提示词的理解细腻，适合追求画面高级感的短片。Known for cinematic quality: realistic physics and lighting with the Ray2 model — for shorts that need a premium look.
最全的 MCP Server 清单（按场景分类）：浏览器、数据库、办公、金融、设计等数十个类别，找现成 MCP 从这份清单开始最省时间。The most comprehensive MCP server list, categorized by scenario — browsers, databases, office, finance and more.
专注 Agent 记忆的中间件：自动从对话中提取、更新、检索事实性记忆，提供托管 API 与自托管两种模式，给任何 Agent 补上「记性」。A memory middleware for agents: automatically extract, update and retrieve factual memories from conversations, hosted or self-hosted.
人人可用的在线设计平台：AI 生图+魔法编辑+海量模板，输入主题自动生成整套 PPT/海报，非设计师做图的第一选择。Design for everyone: AI image generation, magic edit and massive templates — auto-generate full decks and posters from a topic.
微软的企业级 LLM 编排 SDK：C#/.NET 与 Python 双支持，插件化技能、Planner 规划与 Azure 深度集成，是 .NET 技术栈做 AI 应用的标准答案。Microsoft's enterprise SDK for LLM orchestration in C#/.NET and Python, with plugins, planners and deep Azure integration.
主打「Agent 流」的 AI 编辑器：Cascade 自动理解上下文并连续执行多步改动，界面清爽，与 Cursor 齐名的商业 AI IDE。An AI editor built around agentic flow: Cascade keeps context and executes multi-step edits, a peer to Cursor.
谷歌官方机器学习速成课：中文界面、约 15 小时、40+ 可视化交互练习与 pandas 实操，节奏快、覆盖广，非常适合快速建立全局认知。Google's official ML Crash Course: Chinese-supported, ~15 hours, 40+ interactive visual exercises — a fast way to see the whole picture.
桌面端本地模型运行器：图形界面搜索下载 GGUF 模型、调整参数、内置聊天窗口，并提供本地 OpenAI 兼容 API，适合不想碰命令行的用户。A desktop app to discover, download and run local LLMs with a GUI chat interface and a local OpenAI-compatible API server.
拖拽式 LLM 应用构建器：节点画布编排 Agent、RAG 与 API，可自托管、可导出为 API，适合快速交付内部 AI 工具。Drag-and-drop LLM app builder: node canvas for agents, RAG and APIs, self-hostable and exportable as an API.
吴恩达团队的免费短课矩阵：Agent、MCP、RAG、评估、多模态等主题持续更新，每门 1-2 小时可完成，配套 Notebook 实战。Free short courses from Andrew Ng's team covering agents, MCP, RAG, evals and more, each with hands-on notebooks.
高性能 Agent 框架（原 phidata）：支持多模态、记忆、知识库与多 Agent 团队编排，强调低开销与生产可用，内置 40+ 预置 Agent 模板。A high-performance agent framework (formerly phidata) supporting multimodal I/O, memory, knowledge and multi-agent teams with low overhead.
MCP 官方调试器：可视化连接任意 MCP Server，逐个测试 tools/resources/prompts，查看原始 JSON-RPC 消息，开发 Server 的必备伴侣。The official MCP debugger: visually connect to any MCP server, test tools/resources/prompts one by one and inspect raw JSON-RPC messages.
嵌在笔记里的 AI：选中文字即可改写/翻译/总结，能对整个数据库提问，写周报和整理会议记录的效率神器，适合已经用 Notion 的团队。AI built into your notes: rewrite, translate or summarize any selection and query whole databases — ideal for Notion users.
阿里巴巴开源的多 Agent 框架：强调消息传递式协作与容错，内置可视化调试（Studio）与分布式部署能力，中文文档完善。Alibaba's multi-agent framework with message-passing collaboration, fault tolerance, a visual Studio and solid Chinese docs.
收录 Cursor、Devin、v0、Windsurf 等主流 AI 产品的系统提示词，研究顶尖产品如何设计 Agent 指令与工具约束的最佳素材。Leaked system prompts from Cursor, Devin, v0, Windsurf and more — great material for studying how top products prompt agents.
roadmap.sh 的 AI 工程师成长路线：从编程基础、数据/模型概念到 LLM 应用、RAG、Agent、评估部署，一张交互式地图看清学习顺序与技能树全貌。An interactive AI Engineer roadmap on roadmap.sh: from programming fundamentals to LLM apps, RAG, agents and evaluation.
TypeScript 全栈 Agent 框架：Agent、工作流、工具、记忆、RAG、评估一体化，与 Next.js 生态深度集成，前端团队做 AI 应用的顺手选择。A TypeScript-first agent framework with agents, workflows, tools, memory, RAG and evals integrated — a natural fit for Next.js teams.
Supabase 官方 MCP：让 Agent 直接查询表结构、执行 SQL、管理项目配置，做全栈开发时省掉大量来回确认。Supabase's MCP server: inspect schemas, run SQL and manage projects directly from your agent.
DataTalksClub 免费MLOps 实战训练营：实验跟踪、编排、部署、监控、最佳工程实践，9 周做一个端到端项目，从 notebook 走向生产系统的完整训练。A free 9-week MLOps bootcamp: experiment tracking, orchestration, deployment and monitoring — build an end-to-end production ML system.
Agent 可观测性平台：一行代码接入，自动记录每次 LLM 调用、工具执行与成本，支持会话回放、评测与告警，是 Agent 版的「APM」。Agent observability: one line to instrument every LLM call, tool execution and cost, with session replay, evals and alerts — an APM for agents.
国内最大的 SD 模型分享社区：海量国产化Checkpoint/LoRA 模型在线直接生图，免本地部署，找中式风格模型的首选平台。China's largest Stable Diffusion model hub with online generation — find localized LoRA models and run them in the cloud.
让 LLM 在你本机直接写并执行代码的自然语言界面：数据分析、批量文件处理、系统操作，一句话搞定，注意执行前审查代码。A natural-language interface that lets LLMs write and run code locally — data analysis, batch file ops and system tasks.
OpenAI 出品的深度强化学习入门：先讲清 RL 数学符号与算法家族（PPO/SAC 等），再给高质量参考实现，是公认的 Deep RL 第一站。OpenAI's introduction to deep reinforcement learning: clear math notation, algorithm families (PPO, SAC) and high-quality reference implementations.
MemGPT 团队的进化版 Agent 框架：核心是分层记忆管理（核心记忆/归档记忆自动分页），让 Agent 拥有跨会话长期记忆，支持状态持久化与多 Agent。The evolution of MemGPT: tiered memory management that gives agents long-term memory across sessions, with state persistence and multi-agent support.
MCP Server 的「应用商店」：数千个现成 server 一键安装到 Claude/Cursor，也支持托管运行与自建发布，找 MCP 的第一站。An app store for MCP servers: install thousands of ready-made servers into Claude or Cursor in one command.
中文强化学习实践教程：从 MDP 基础到 DQN、策略梯度、SAC，理论与代码交替推进，每章附可运行 Jupyter，适合中文读者入门 RL。A Chinese hands-on RL tutorial from MDP basics to DQN, policy gradients and SAC, with runnable notebooks per chapter.
开源 LLM 可观测与评估平台：基于 OpenTelemetry 追踪 LLM/Agent/工具调用链，配套数据集驱动的评估体系，可自托管、厂商中立。Open-source LLM observability: OpenTelemetry-based tracing for LLMs, agents and tools, plus dataset-driven evals — self-hosted and vendor-neutral.
主打创意特效的 AI 视频工具：灵魂出窍、物体爆炸、捏捏脸等病毒式模板一键套用，社交媒体玩梗和趣味内容的快乐源泉。AI video with viral creative effects — squeeze, explode, inflate and more one-click templates for social content.
引爆 2023 年 Agent 热潮的开山项目，如今已重构为可视化 Agent 构建平台（低代码积木 + 市场），适合搭建自动化任务流程。The project that started the 2023 agent wave, now rebuilt as a visual low-code agent builder with a block marketplace.
号称「100 行」的极简 Agent 工作流框架：只抽象节点与共享存储两个概念，其余全部交给提示词，配套大量「从零实现 XX Agent」教程，极适合学习原理。A minimalist agent workflow framework in ~100 lines: only nodes and shared store abstractions, plus tons of build-from-scratch tutorials.
官方文件系统 MCP Server：让 AI 安全地读写指定目录的文件、列目录、搜索内容，既是现成工具也是学习 Server 实现的官方范例。The official filesystem MCP server: safe read/write within allowed directories — both a ready-to-use tool and a reference implementation.
微软的图谱增强 RAG 方案：先抽取实体关系构建知识图谱，再结合社区摘要回答全局性问题，适合大文档集的「总结型」问答。Microsoft's graph-based RAG: build a knowledge graph then answer global, summarization-style questions over large corpora.
把软件公司搬进多 Agent 系统：CEO/CTO/程序员/测试员角色协作，输入一句话需求输出可运行小程序，中文社区活跃，适合研究角色编排。A virtual software company of agents: CEO, CTO, programmers and testers collaborate to turn one-line requirements into runnable apps.
Google 官方 Gemini 提示词画廊：按任务分类的成品 prompt（多模态、代码、长上下文等），每个附可直接运行的示例，学官方写法的捷径。Google's official Gemini prompt gallery: task-classified production prompts with runnable examples across multimodal and long-context scenarios.
社区维护的 Claude Skills 精选清单，收录大量第三方技能包与开发资源，找 Skill 灵感的首选入口。A curated community list of Claude Skills, third-party skill packages and developer resources.
最流行的 Python MCP Server 开发框架，几行装饰器即可把函数发布为 MCP 工具，开发效率极高。The popular Pythonic framework for building MCP servers — expose functions as tools with a few decorators.
功能最全的开源 LLM 前端：支持 Ollama/各大 API、RAG、工具调用与多用户管理，Docker 一键部署本地 AI 工作台。The most feature-rich open-source LLM UI: Ollama/API support, RAG, tools and multi-user management, one-click Docker deploy.
面向 AI 编码的任务编排器：把 PRD 拆解成依赖任务清单，逐个驱动 Claude/Cursor 执行并跟踪状态，让「一句话需求」变成可管理的开发流。A task orchestrator for AI coding: split PRDs into dependent tasks and drive Claude/Cursor through them with status tracking.
Notion 官方 MCP Server：AI 可搜索、读取、创建和更新 Notion 页面与数据库，把团队知识库接进 AI 工作流。Notion's official MCP server: search, read, create and update Notion pages and databases from AI clients.
把软件公司「角色分工」搬进多 Agent 系统：产品经理、架构师、工程师协作，输入一句话需求输出代码与文档。A multi-agent system simulating a software company: PM, architect and engineer roles turn one-line requirements into code and docs.
面向 Agent 的工具集成平台：250+ 应用（GitHub、Slack、Notion 等）的开箱即用工具与鉴权管理，支持 MCP 输出。Tool integration platform for agents: 250+ ready-made app tools (GitHub, Slack, Notion...) with auth management and MCP support.
最经典的 Prompt 提示词合集（100k+ Star），收录上百种角色化提示词，学写 Prompt 的第一课教材。The classic prompt collection (100k+ stars) with hundreds of role-based prompts — everyone's first prompt engineering textbook.
社区维护的 MCP Server 导航站：按场景与客户端分类检索，附带安装说明，适合快速横向对比同类 server。A community-curated MCP server directory with scenario-based browsing and install notes.
官方 SQLite MCP Server：AI 可查询表结构、执行 SQL、分析数据，本地数据分析场景零门槛上手，也是数据库类 Server 的参考实现。The official SQLite MCP server: schema inspection, SQL execution and data analysis — a reference for database MCP servers.
Vercel v0 平台的提示词示例集：展示如何用自然语言描述 UI 需求生成 React/Tailwind 组件，是学习「结构化描述界面」的最佳样本。Prompt examples from Vercel's v0: learn how to describe UI requirements in natural language to generate React/Tailwind components.
微软开源的多 Agent 对话框架，让多个 Agent 通过对话协作完成复杂任务，支持人机混合参与，学术与工业界应用广泛。Microsoft's open-source framework where multiple agents converse to solve complex tasks, with human-in-the-loop support.
原 OpenDevin，开源的 AI 软件工程师 Agent：能改代码、跑命令、浏览网页、提交 PR，在 SWE-bench 上表现领先。Formerly OpenDevin: an open-source AI software engineer that edits code, runs commands, browses and opens PRs.
为 LLM 设计的网页抓取 API：整站爬取并转为干净的 Markdown，自带 MCP Server，是 Agent 联网获取资料的热门选择。Web scraping API built for LLMs: crawl sites into clean Markdown, with an official MCP server for web-connected agents.
Brex 工程团队的提示词工程指南：从基本原则到部署、评测与维护的全生命周期视角，比入门教程更贴近生产环境的复杂现实。Brex's prompt engineering guide covering the full lifecycle — principles, deployment, evaluation and maintenance — closer to production reality.
AI 驱动的多平台舆情监控工具：聚合 GitHub/HN/Reddit 等平台热点，中文摘要日报推送，适合关注技术趋势的个人或团队每日追踪。AI-powered public opinion and trend monitor aggregating GitHub, HN and Reddit with Chinese digest reports for daily tech tracking.
以「角色扮演」为核心的多 Agent 框架：为每个 Agent 设定角色、目标与工具，像组建团队一样编排 AI 协作，上手简单、社区活跃。Role-playing multi-agent framework: assign roles, goals and tools to each agent and orchestrate them like a team.
可自托管的工作流自动化平台，内置大量应用集成节点与 AI Agent 能力，是把 Agent 接入真实业务系统的胶水层。Self-hostable workflow automation with hundreds of integrations and native AI agent nodes — the glue between agents and real business systems.
让 AI Agent 直接「使用浏览器」的 Python 库：视觉+DOM 双通道理解页面，网页操作任务成功率业界领先。Python library that lets agents drive a browser visually and via DOM — state-of-the-art on web task benchmarks.
Anthropic 官方提示词库：办公、写作、编程、分析等多场景高质量 Prompt，附元提示词（Prompt 生成 Prompt）技巧。Anthropic's official prompt library: high-quality prompts for office, writing, coding and analysis, plus meta-prompting techniques.
MCP Server 的包管理器：像 npm 管包一样安装/卸载/更新各客户端的 MCP Server，一处配置多客户端同步，告别手改 JSON。A package manager for MCP servers: install, uninstall and sync MCP servers across clients from one CLI, no more hand-editing JSON.
开源 AI 聊天应用框架：多模型接入、插件生态、知识库与 Agent 市场，界面精美，适合自建个人 AI 助手站。Open-source AI chat framework with multi-model support, plugins, knowledge bases and an agent marketplace.
Slack MCP Server：AI 可读取频道消息、发消息、回复线程，用于「帮我总结今天频道讨论」这类协作自动化场景。A Slack MCP server for reading channels, posting messages and replying in threads — enabling collaboration automations.
字节跳动的零代码 AI Bot 搭建平台：拖拽式工作流、插件市场、知识库与长期记忆，可一键发布到飞书、微信等渠道。ByteDance's no-code platform for building AI bots with workflows, plugin marketplace, knowledge bases and one-click publishing.
开源 VSCode 编程 Agent 插件：规划-执行-审查闭环，支持任意模型与 MCP 工具，可全程人工确认，安全可控。Open-source VSCode coding agent plugin with plan-act-review loops, any-model support and human approval at every step.
Hugging Face 官方免费 MCP 课程：从协议原理到开发自己的 MCP Server 与 Agent 集成，配套实战单元。Hugging Face's free official MCP course: protocol fundamentals to building your own MCP servers and agent integrations.
微软官方的浏览器自动化 MCP Server，让 Agent 可以操作网页、填表、截图、跑端到端测试，网页 Agent 开发必备。Microsoft's official MCP server for browser automation: let agents browse, fill forms, screenshot and run E2E tests.
AI 原生代码编辑器的代表产品：代码库级理解、Tab 智能补全、Agent 模式自动完成多文件改动，开发者口碑极佳。The flagship AI-native code editor: repo-level understanding, smart tab completion and an agent mode for multi-file changes.
OpenAI 官方示例代码库：Agent 编排、函数调用、RAG、评估等实战 Notebook，跟随 API 更新持续迭代。OpenAI's official examples: agent orchestration, function calling, RAG and evaluation notebooks, kept in sync with the API.