AI & ML
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
-
ahmedesmail07 Bundle QA VerifyDeep verification of a specific behavior, bug, or fix using Playwright MCP. Focused single-issue testing with thorough state inspection (DOM, console, network, JS state). Examples: "/qa-verify auto-save fires on blur", "/qa-verify fix for #123", "/qa-verify login redirect"
-
ahmedesmail07 Bundle QA ExecuteExecute QA test cases using Playwright MCP browser tools. Navigates, interacts, validates, screenshots. Works with any web app. Give it a URL and test cases (or let it discover them). Examples: "/qa-execute http://localhost:3000", "/qa-execute http://localhost:9174 plan:test-plan.md"
-
tashangkd Bundle Skill DesignerSkill/Agent/Rule 开发全流程双视角导引者。新建或重构任何 Skill/Agent/Rule 时激活。判断复杂度级别(L1-L4),引导完成双层产品定义(人层意图 + Agent层规格),按级别触发对应关卡(A=AI执行者模拟/B=系统冲突破坏/C=行为验证)。触发词:「新建Skill」「设计一个Skill」「要写一个新的Skill」「新建Agent」「设计子智能体」「新建Rule」「设计规则」。
-
nihao555-hub Skill Agent ArchitectAI Agent 架构设计与系统组织专家 - 掌握 Claude Code的整体架构、模块组织、依赖管理和设计模式
-
nihao555-hub Skill Prompt EngineerAI Agent 提示词工程专家 - 系统提示设计、动态注入、多 Agent 协调、fork语义、工具描述生成
-
nihao555-hub Skill Skill DeveloperAI Agent 技能开发专家 - Bundled Skill注册、文件提取机制、参数处理、钩子系统、技能搜索和动态加载
-
nihao555-hub Skill Tool IntegratorAI Agent 工具集成专家 - MCP 协议、OAuth 认证、工具分类、权限控制、活动描述生成
-
nihao555-hub Skill Context EngineerAI Agent 上下文工程专家 - Git 上下文、内存层级、会话状态追踪、@include指令处理、PII 保护和智能缓存策略
-
nihao555-hub Skill Harness EngineerAI Agent Harness工程专家 - 任务编排、进度追踪、状态管理、磁盘引导、优雅关闭
-
nihao555-hub Skill Debug DiagnosticianAI Agent 调试诊断专家 - 会话调试、日志分析、错误追踪、堆栈分析、诊断跟踪服务
-
nihao555-hub Skill Performance OptimizerAI Agent 性能优化专家 - 启动剖析、懒加载、缓存策略、内存管理、并行执行、性能监控
-
nihao555-hub Skill Verification EngineerAI Agent 验证测试专家 - 变更验证、自动化测试、质量保障、回归检测、测试覆盖率分析
-
tashangkd Bundle Skill Rule修改 Skill / Rule / Agent 文件的操作规范。当 AI 或用户需要新增、修改、删除 .cursor/skills/ 或 .cursor/rules/ 或 .cursor/agents/ 中的任何文件时触发。修改前必须遵循本规范,未经本规范允许不可直接改写 Skill/Rule/Agent。
-
tashangkd Skill Expert Bootstrap按需培养领域专家子智能体。通过调研→整理→自问自答反思→专家配置输出四阶段,将任何领域的知识转化为可 dispatch 的专家 Agent。Orchestrator 在识别到「现有审查能力不足」时触发。触发词:「需要X领域的专家」「这个任务需要X领域的深度审查」「培养专家」「我们没有X方面的专家」「expert-bootstrap」「build expert」「专家培养」。
-
tashangkd Skill Generate AI Memory Prompt根据当前科研数字分身状态,生成结构化提示词,供用户提交给 ChatGPT 等带记忆功能的 AI,从中提取与画像构建相关的信息。不询问用户使用哪个 AI 工具,提示词对所有平台通用。
-
tashangkd Bundle Skill System Health CheckSkill体系自洽性审查 Skill(由用户主动触发,不被其他组件调用)。主动检查整个Skill/Agent/Rule体系的内部一致性:触发词冲突、孤立组件、版本漂移、SKILL-INDEX与实际文件不一致、alwaysApply Rule叠加矛盾等。对应产品侧的 project-closeout。注意:与 skill-system-destroyer(子智能体,只读审查新组件冲突)不同——本 Skill 审查整个体系的历史一致性。触发词:「检查Skill体系」「Skill体系有没有问题」「Skill体系自检」「Skill体系自洽性」「Skill健康检查」。
-
nihao555-hub Skill Error Handling SpecialistAI Agent 错误处理专家 - 统一错误类型、错误传播、降级策略、错误恢复、容错设计
-
nihao555-hub Skill State Management EngineerAI Agent 状态管理专家 - 不可变状态、状态订阅、状态持久化、乐观更新、撤销重做
-
nihao555-hub Skill MCP Integration SpecialistMCP 集成专家 - MCP 服务器开发、传输协议配置、OAuth 认证、工具注册、资源管理
-
nihao555-hub Skill Security Compliance OfficerAI Agent 安全合规专家 - 权限分级、PII 保护、策略执行、安全审计、合规检查、denial tracking
-
nihao555-hub Skill Testing Quality Assurance EngineerAI Agent 测试质量保证专家 - 单元测试、集成测试、E2E 测试、测试覆盖率、质量门禁
-
manojsukumaran-ops Bundle Build MCP ServerGuide enterprise teams through building a production-ready MCP (Model Context Protocol) server using FastMCP (Python). Use when users want to "build an MCP server", "create MCP tools", "convert a REST API to MCP", "expose an API as MCP tools", "scaffold an MCP server", or "add MCP support to an existing service". Enforces TDD (tests before implementation), OAuth middleware for inbound auth validation, structured tool logging, and resource-level tool grouping as enterprise best practices.
-
tashangkd Bundle Skill Library Search全量搜索外部 Skill/Agent/Rule 仓库,必须执行7维度协议,禁止以少于7维度的搜索声称「全量完成」。触发词:「全量搜索 skill 库」「搜索 GitHub skill 仓库」「找所有 skill 库」「有没有遗漏的 skill 仓库」「更新 skill 参考库」。
-
tashangkd Bundle Skill Capture Closure任务结束后的统一复盘入口,同时覆盖「修改」(识别并修复 B/K 对象缺陷)和「沉淀」(将新经验/洞见写入对应 B/K 对象)两个方向。主分类轴:B-object(Skill/Rule/Agent)vs K-object(原则/框架/规范/项目规格/碎片),操作类型:modify(修改已有)或 create(新建沉淀)。触发词:「复盘一下」「做个复盘」「这次踩坑了」「这个经验值得记」「跑通了」「总结经验」「沉淀一下」「记录一下这次」。注意:项目级批量复盘请触发 project-retrospective。
-
chuanzige Bundle Worker Prompt CraftUse when assigning work to a worker or subagent that cannot see the main conversation, especially when the prompt must be self-contained, precise, and resistant to ambiguity or lazy follow-up delegation
-
chuanzige Bundle Coordinator OrchestratorUse when coordinating multi-step work across workers or subagents, especially when research, implementation, and verification should be split while the main agent stays responsible for synthesis and user communication
-
blackblazexxx Bundle Visual Promptv4.0 Meta-Level Visual DNA Prompt Engine (Data-Driven Visuals)
-
yikangai-pixel Bundle Project Harness BlueprintUse when the user wants to turn a new project idea into a reusable harness blueprint for agent-assisted delivery. This skill is for project setup before implementation: repo/docs foundation, agent execution workflow, and verification design. Use for MVP planning, internal tools, new products, or when the user asks how to describe a project so Codex can scaffold the operating model.
-
thu-nmrc Bundle Harness 24hA production-grade framework for long-running, autonomous agents based on Harness Engineering principles. Features three-layer self-healing memory, circuit breaker protection, KAIROS dream-mode consolidation, and multi-agent coordination. When a user describes a task idea, Agent automatically initializes the workspace, fills all template files, sets up the cron schedule, and begins execution immediately.
-
maruko912 Bundle Joyspace Create在JoySpace中创建新文档并输入标题和内容。通过Chrome DevTools MCP操作浏览器完成。
-
chunkydotdev Skill Inbox ProvidersChoose and configure an agent-native inbox provider for AI agents that send and receive email. Use when giving an AI agent its own email address, setting up two-way agent email communication, comparing agent inbox platforms, or adding guardrails to autonomous email workflows.
-
chenluda Bundle Claudecode Architecture Design GuideExtract and apply architecture guidance from the Claude Code engineering system to other projects. Use when doing system architecture design, technical strategy, building agent/tool/plugin platforms, designing permission models, streaming pipelines, state machines, or phased migration planning.
-
kadaliao Bundle Memory Type System结构化记忆类型系统 — 四类型分类 + frontmatter 规范 + 写入/召回/漂移防护全流程。适用于所有 Agent 的长期记忆管理。解决记忆混乱、召回不准、膨胀失控问题。灵感源自 Claude Code Memory Type Taxonomy(memoryTypes.ts)。
-
kadaliao Bundle Smart Memory Guard智能记忆管理 — 防膨胀、防漂移、分类写入。解决记忆文件无限增长、过时记忆误导行动、什么都记导致噪音太多的问题。适用于所有 OpenClaw Agent。触发条件:(1) 写入 MEMORY.md 前自动执行分类检查 (2) 基于记忆行动前验证漂移 (3) heartbeat 时检查膨胀 (4) 被主人说"记忆太乱了"时。灵感源自 Claude Code Memory 系统设计。
-
kadaliao Bundle Worker Prompt CraftWorker/子代理 Prompt 编写指南 — 写出让子代理一次做对的 prompt。解决"子代理看不到上下文"、"指令模糊导致返工"、"委托理解而非综合"的问题。适用于:(1) 使用 sessions_spawn/sessions_send 时 (2) 调度 CC 子任务时 (3) 任何需要给 AI 子代理写指令的场景。灵感源自 Claude Code Coordinator Prompt Guidelines。
-
kadaliao Bundle Lightweight Explorer轻量探索模式 — 用最少 token 完成代码搜索和调研。核心理念:只读搜索不需要加载全部工具和项目规则。适用于:(1) 搜索代码库找文件/函数 (2) 调研项目架构 (3) 多文件分析 (4) 回答"代码里哪里有 X"类问题。灵感源自 Claude Code Explore Agent 设计。
Frequently asked questions
What are AI & ML agent skills?
AI & ML agent skills cover the machine-learning workflow itself: writing and evaluating prompts, building RAG pipelines, running evals, and wiring up model APIs. Each one is a SKILL.md file your agent loads on demand, so the know-how travels across Claude Code, Cursor, and 60+ agents.
Which AI & ML skills are most installed?
Popular AI & ML skills on SkillMD right now include qa-verify, qa-execute, skill-designer. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do AI & ML skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.