Cursor · Windsurf · Claude Code · GitHub Copilot · Codex · MCP

Copy-ready Rules, Roles, Skills and Hooks for AI Coding Agents

Browse practical AI coding assets for Cursor, Windsurf, Claude Code, GitHub Copilot, Codex and MCP workflows. Find reusable rules, roles, skills, hooks and workflows for building, debugging, reviewing, refactoring, testing and deploying software with AI.

Browse AI Coding Assets Quick Hooks Submit an Asset

Browse by Tool

Filter copy-ready assets for your AI coding agent.

CursorRules, roles, skills & hooks WindsurfCascade rules & prompts Claude CodeProject instructions & roles GitHub CopilotInstructions & prompts CodexAgent system prompts MCPConfigs & workflows

Browse by Asset Type

Rules for standards, roles for persona, skills for tasks, hooks for quick prompts.

Rules Roles Skills Hooks Workflows MCP Configs

Browse by Goal

Jump to assets for common developer workflows.

Build SaaS MVP Fix Bugs Review Code Refactor Project Add Tests Deploy App Improve UI Secure App Write Docs

How it works

Choose an asset

Browse by tool, asset type, category, framework, or goal — or pick a popular template above.

Copy the prompt

Open any asset and click Copy to grab the full rule, role, skill, or hook text.

Paste into your agent

Use in Cursor, Windsurf, Claude Code, GitHub Copilot, Codex, or MCP — adapt as needed for your workflow.

Quick Hooks for AI Coding Agents

Short prompts you can paste into Cursor, Windsurf or Claude Code before starting a task.

AI Coding Assets Directory

Filter by tool, asset type, category, and framework to find the right rules, roles, skills, hooks, workflows, or MCP configs.

Framework

WINDSURF RULES BACKEND / API

Python Web Scraping Rules

Ethical, robust, and maintainable Python data collection and parsing.

CURSOR RULES DATABASE / ORM

Supabase Database Rules

Row Level Security, Edge Functions, and Postgres best practices.

WINDSURF RULES BACKEND / API

Go Microservice API Rules

RESTful design, error handling, and high-performance Go backend patterns.

WINDSURF RULES BACKEND / API

FastAPI Backend Rules

Async routes, Pydantic validation, and OpenAPI documentation for Python APIs.

CURSOR RULES AI AGENT / RAG

MCP Server Development Rules

Build Model Context Protocol servers with typed tools, resources, and secure transport.

CURSOR RULES DOCUMENTATION / PRODUCTIVITY

Cursor Agent & Skills Rules

Guide Cursor Agent behavior, Skills usage, and safe autonomous coding workflows.

CURSOR RULES AI AGENT / RAG

Vercel AI SDK Rules

Streaming chat UIs, tool calling, and server-side AI routes with Vercel AI SDK.

CURSOR RULES TESTING / QA

Playwright E2E Testing Rules

Reliable end-to-end tests with Playwright — selectors, fixtures, and CI patterns.

WINDSURF RULES TESTING / QA

Vitest Unit Testing Rules

Fast unit and component tests with Vitest, mocks, and coverage targets.

CURSOR RULES DATABASE / ORM

Prisma ORM Rules

Schema design, migrations, typed queries, and Prisma Client best practices.

WINDSURF RULES BACKEND / API

Node.js Express API Rules

REST APIs with Express — middleware, validation, error handling, and structure.

CURSOR RULES DEVOPS / DEPLOYMENT

GitHub Actions CI/CD Rules

Reusable workflows, caching, matrix builds, and secure CI pipelines.

CURSOR RULES CODE QUALITY / SECURITY

Auth.js / NextAuth Rules

Authentication with Auth.js — providers, sessions, middleware, and RBAC.

WINDSURF RULES CODE QUALITY / SECURITY

Zod Schema Validation Rules

Runtime validation, inferred types, and API boundary schemas with Zod.

CURSOR RULES FRONTEND / UI

SvelteKit Frontend Rules

SvelteKit routing, load functions, forms, and progressive enhancement.

WINDSURF RULES DATABASE / ORM

PostgreSQL Database Rules

Schema design, indexing, queries, migrations, and Postgres-specific features.

WINDSURF RULES CODE QUALITY / SECURITY

ESLint & Prettier Rules

Lint and format setup — flat config, CI integration, and team conventions.

CURSOR RULES DOCUMENTATION / PRODUCTIVITY

Turborepo Monorepo Rules

Monorepo structure, turbo pipelines, shared packages, and caching.

CURSOR RULES FRONTEND / UI

shadcn/ui Component Rules

Accessible UI components with shadcn/ui, Radix primitives, and Tailwind.

WINDSURF RULES AI AGENT / RAG

Prompt Engineering Rules

Effective system prompts, few-shot examples, and structured outputs for LLMs.

CURSOR RULES FULLSTACK

Next.js App Router + TypeScript Rules

Rules for building clean, scalable Next.js App Router projects with TypeScript, server components, API routes, and reusable architecture.

CURSOR RULES FRONTEND / UI

React + TypeScript Component Rules

Rules for writing maintainable React components with TypeScript, clear props, reusable hooks, and clean component structure.

WINDSURF RULES FRONTEND / UI

Tailwind CSS + shadcn/ui Rules

Rules for building modern UI using Tailwind CSS and shadcn/ui while keeping components clean, accessible, and consistent.

CURSOR RULES DATABASE / ORM

Supabase Auth + RLS Security Rules

Rules for building secure Supabase apps with authentication, row level security, policies, and safe database access.

WINDSURF RULES DEVOPS / DEPLOYMENT

Cloudflare Workers + Pages Rules

Rules for building and deploying lightweight apps on Cloudflare Workers and Pages with edge-first architecture.

CURSOR RULES MOBILE

Expo React Native Mobile App Rules

Rules for creating clean Expo React Native apps with reusable screens, navigation, state management, and mobile-friendly structure.

WINDSURF RULES BACKEND / API

Python FastAPI Backend Rules

Rules for building clean FastAPI backends with typed routes, Pydantic models, validation, services, and error handling.

CURSOR RULES CODE QUALITY / SECURITY

AI Code Review + Security Rules

Rules for asking AI coding agents to review code for bugs, security risks, maintainability, performance, and edge cases.

CURSOR ROLES FULL-STACK

Senior Full-Stack Developer Role

A senior engineering role prompt for building production-ready full-stack features with clear planning, clean architecture, and safe implementation.

CURSOR ROLES FULL-STACK

Next.js Architect Role

A Next.js architecture role for App Router, server components, API routes, and production-ready structure.

CURSOR ROLES FRONTEND

TypeScript Expert Role

A TypeScript role for safer types, better interfaces, and fewer runtime errors.

CURSOR ROLES TESTING

AI Code Reviewer Role

A code review role for finding bugs, maintainability issues, security risks, and edge cases.

CURSOR ROLES DEBUGGING

Debugging Expert Role

A debugging role that focuses on root cause analysis before changing code.

CURSOR ROLES SECURITY

Security Engineer Role

A security role for authentication, authorization, validation, secrets, and data exposure.

CURSOR ROLES SAAS

SaaS Product Engineer Role

A product-focused engineering role for building SaaS MVP features quickly and safely.

CURSOR SKILLS SAAS

Build SaaS MVP Skill

Build a simple SaaS MVP with authentication, dashboard, database, and clean structure.

MCP WORKFLOWS AI AGENT

MCP Server Setup Workflow

Step-by-step workflow for setting up a Model Context Protocol server and connecting it to AI coding agents.

MCP MCP CONFIGS AI AGENT

GitHub MCP Config

A practical MCP configuration concept for connecting an AI coding agent to GitHub repositories, issues, pull requests, and code review workflows.

MCP MCP CONFIGS AI AGENT

Filesystem MCP Config

A filesystem MCP setup concept for giving an AI coding agent structured access to local project files.

MCP MCP CONFIGS TESTING

Browser MCP Config

A browser MCP setup concept for web inspection, UI testing, debugging, and browser-based workflows.

MCP MCP CONFIGS DATABASE

Database MCP Config

A database MCP setup concept for helping AI agents understand schemas, queries, migrations, and data relationships.

MCP MCP CONFIGS FRONTEND

Figma MCP Config

A Figma MCP setup concept for connecting design context to frontend implementation.

CLAUDE CODE ROLES REFACTORING

Claude Code Refactoring Role

A Claude Code role for safely refactoring existing projects without changing behavior.

CODEX ROLES TESTING

Codex Code Review Role

A Codex role for reviewing code quality, correctness, maintainability, and risk.

Popular AI Coding Use Cases

Cursor Rules for Next.js

Production-ready Cursor rules for Next.js App Router, server components, and TypeScript. Copy templates for full-stack features, API routes, and performance best practices.

Cursor Rules for React

Reusable React frontend rules and roles for components, hooks, and accessible UI. Ideal for teams standardizing AI-assisted React development.

Cursor Rules for TypeScript

TypeScript rules and expert roles for stricter types, clearer interfaces, and fewer runtime errors across frontend and backend codebases.

AI Code Review Prompts

Senior reviewer roles, PR review skills, and one-line hooks for catching bugs, security risks, and maintainability issues before merge.

Debugging Prompts for Cursor

Debugging expert roles and step-by-step skills that force root cause analysis before code changes — fewer random fixes, faster resolution.

Playwright Testing Prompts

E2E testing rules, automation engineer roles, and Playwright skills for stable browser tests and CI-ready user journey coverage.

MCP Configs for AI Agents

GitHub, filesystem, browser, database and Figma MCP configuration concepts for connecting AI coding agents to external tools and project context.

Claude Code Roles & Skills

Copy-ready Claude Code roles, skills and hooks for refactoring, debugging, codebase explanation and safe AI-assisted development workflows.

GitHub Copilot Prompts

PR review prompts, test generation skills and issue-to-code workflows for GitHub Copilot Chat and copilot-instructions.md.

AI Coding Hooks

Short pre-task hooks for planning, debugging, refactoring, security audits and deployment prep — paste before starting any coding session.

Cursor Assets Directory

Browse Cursor rules, roles, skills, hooks and workflows for Agent and Chat — from Next.js and React to testing, DevOps and security.

Windsurf Skills & Workflows

Cascade-ready Windsurf skills, rules and workflows for full-stack, backend, testing and AI agent development.

MCP Workflows for Developers

Model Context Protocol setup workflows and MCP server rules for connecting tools, resources, and agents in Cursor and other AI coding environments.

FAQ

Where do Cursor rules go?

Put project rules in .cursor/rules/ at your repo root. Use .mdc files (recommended) or Markdown — for example .cursor/rules/*.mdc. Cursor applies them automatically in Agent and Chat. See our Cursor Agent & Skills rules for workflow tips.

Where do Windsurf / Devin Desktop rules go?

For workspace rules, create Markdown files under .devin/rules/*.md (preferred) or legacy .windsurf/rules/*.md at your repo root. Legacy .windsurfrules files are still read. You can also manage rules from Cascade Customizations → Rules.

What about AGENTS.md?

AGENTS.md is another supported project instruction format — a repo-level brief describing architecture, commands, and workflow. It complements tool-specific rules files; use both when your project needs a high-level map plus detailed coding standards.

How do I use an asset from AI Agent Dock?

Find an asset in the directory, open its detail page, click Copy (the button label matches the asset type — Rule, Role, Skill, Hook, Workflow, or MCP Config), then paste into your editor. Each asset has a permanent URL you can bookmark or share (e.g. /rules/fastapi-backend).

Can I edit assets after copying?

Yes. Assets are starting templates — remove sections you do not need, add project-specific conventions, and keep total length reasonable so you do not overrun the AI context window.

Are these assets free to use?

Yes. Assets on this site are original templates or properly attributed community submissions. Most are MIT-licensed unless noted otherwise on the detail page. Check the License line on each card or detail page.

How do I submit my own asset?

Visit Submit an AI Coding Asset for guidelines, then send your asset by email.

Trust and licensing

Assets on AI Agent Dock are original templates or properly attributed community submissions. Most assets are MIT-licensed unless noted otherwise on the detail page. See About for attribution details.