AI-Assisted Development Toolkit

A comprehensive template repository providing AI instruction files, prompts, and rules for AI-assisted development workflows. This toolkit helps teams establish consistent AI coding practices across projects.

Purpose

This repository serves as a template source for AI development instructions that can be copied into your projects. It provides:

• Cursor IDE rules (.cursor/rules/) for AI behavior configuration
• Cursor commands (.cursor/commands/) for custom AI workflows
• GitHub Copilot instructions (.github/copilot-instructions.md)
• GitHub prompts (.github/prompts/) for reusable prompt templates
• AGENTS.md files for AI agent context
• Skills recommendations from skills.sh for enhanced AI capabilities

Repository Structure

/
├── src/
│   ├── repo/           # Templates for single repositories
│   │   ├── .cursor/    # Cursor IDE configuration
│   │   ├── .github/    # GitHub Copilot instructions
│   │   ├── .agents/    # Pre-installed universal skills
│   │   │   └── skills/ # Bundled skills (TDD, debugging, etc.)
│   │   ├── AGENTS.md   # AI agent context template
│   │   └── README.md   # Project README template
│   │
│   └── monorepo/       # Templates for monorepos
│       ├── .cursor/    # Root-level Cursor config
│       ├── .github/    # Root-level GitHub config
│       ├── .agents/    # Pre-installed universal skills
│       │   └── skills/ # Bundled skills (TDD, debugging, etc.)
│       ├── apps/       # App-specific templates
│       ├── packages/   # Package-specific templates
│       └── AGENTS.md   # Monorepo agent context
│
├── .cursor/            # This toolkit's own Cursor config
├── .github/            # This toolkit's own GitHub config
└── AGENTS.md           # This toolkit's agent context

Usage

For Single Repositories

Step 1: Copy the templates

cp -r src/repo/.cursor /path/to/your/project/
cp -r src/repo/.github /path/to/your/project/
cp -r src/repo/.agents /path/to/your/project/
cp src/repo/AGENTS.md /path/to/your/project/

The .agents/skills/ folder contains pre-installed universal skills that are ready to use immediately (see Bundled Universal Skills).

Step 2: Run the Bootstrap Command

Open your project in Cursor IDE or VS Code with GitHub Copilot, then run:

• Cursor IDE: Use the /bootstrap command
• GitHub Copilot: Use the bootstrap.prompt.md prompt (/bootstrap)

The bootstrap process will:

1. Analyze your project's structure, dependencies, and configuration files
2. Infer template variables (framework, language, package manager, etc.)
3. Prompt you for any values it couldn't detect
4. Automatically update all {{PLACEHOLDER}} values in your AI instruction files

Step 3: Review and Customize

After bootstrapping, review the generated files and add any project-specific patterns to AGENTS.md.

For Monorepos

Step 1: Copy the templates

cp -r src/monorepo/.cursor /path/to/your/project/
cp -r src/monorepo/.github /path/to/your/project/
cp -r src/monorepo/.agents /path/to/your/project/
cp src/monorepo/AGENTS.md /path/to/your/project/
# Optionally copy app/package templates
cp -r src/monorepo/apps/app-template /path/to/your/project/apps/
cp -r src/monorepo/packages/package-template /path/to/your/project/packages/

The .agents/skills/ folder contains pre-installed universal skills that are ready to use immediately (see Bundled Universal Skills).

Step 2: Run the Bootstrap Command

Open your project in Cursor IDE or VS Code with GitHub Copilot, then run:

• Cursor IDE: Use the /bootstrap command
• GitHub Copilot: Use the bootstrap.prompt.md prompt (/bootstrap)

The bootstrap process will:

1. Analyze your monorepo workspace, including all apps and packages
2. Detect frameworks, ports, and deployment platforms for each app
3. Infer package scopes and descriptions
4. Prompt you for any missing information
5. Create AGENTS.md files for each app with framework-specific patterns
6. Update all {{PLACEHOLDER}} values throughout the monorepo

Step 3: Review and Customize

After bootstrapping:

• Review root AGENTS.md for overall project context
• Check each apps/*/AGENTS.md for app-specific patterns
• Add any custom conventions or patterns unique to your project

Placeholder Reference

Templates use {{PLACEHOLDER}} syntax for values you need to customize. The bootstrap command will automatically detect and fill most of these values.

Common Placeholders

Placeholder	Description	Example
{{PROJECT_NAME}}	Your project name	my-awesome-app
{{PROJECT_DESCRIPTION}}	Brief project description	A web app for...
{{LANGUAGE}}	Primary language	TypeScript, PHP, C#, Python, Ruby, Go, Rust, Java
{{FRAMEWORK}}	Primary framework	Next.js, Laravel, Blazor, Django, Rails, Spring Boot
{{PACKAGE_MANAGER}}	Package manager	npm, composer, dotnet, pip, bundler, go, cargo, maven
{{FILE_EXTENSION}}	File extension	ts, php, cs, py, rb, go, rs, java
{{DEV_PORT}}	Development server port	3000, 8000, 5000
{{DEPLOY_PLATFORM}}	Deployment target	Vercel, AWS, Azure, Heroku
{{DATABASE}}	Database/ORM	Prisma, Eloquent, EF Core, SQLAlchemy, ActiveRecord
{{TEST_FRAMEWORK}}	Testing framework	Vitest, PHPUnit, xUnit, pytest, RSpec, JUnit
{{ISSUE_TRACKER}}	Issue tracking system	Jira, Linear, GitHub Issues
{{DEFAULT_BRANCH}}	Default git branch	main, master
{{RUNTIME_VERSION}}	Runtime version	Node 20, PHP 8.3, .NET 8, Python 3.12
{{REPO_URL}}	Repository URL	https://github.com/org/repo
{{SRC_DIR}}	Source directory	src, app, lib
{{TEST_DIR}}	Test directory	tests, test, spec
{{STYLING}}	Styling solution	Tailwind, CSS Modules, SCSS
{{NODE_VERSION}}	Node.js version	20, 22
{{LICENSE_TYPE}}	License identifier	MIT, Apache-2.0, ISC
{{SRC_STRUCTURE}}	Source folder structure	components/, lib/, utils/

Command Placeholders

Placeholder	Description	Example
{{INSTALL_COMMAND}}	Install dependencies	npm install, composer install, dotnet restore
{{DEV_COMMAND}}	Run dev server	npm run dev, php artisan serve, dotnet watch
{{BUILD_COMMAND}}	Build for production	npm run build, dotnet publish
{{TEST_COMMAND}}	Run tests	npm test, ./vendor/bin/phpunit, dotnet test
{{LINT_COMMAND}}	Run linting	npm run lint, ./vendor/bin/pint

Monorepo-Specific Placeholders

Placeholder	Description	Example
{{APP_NAME_1}}	First app name	web, frontend
{{APP_NAME_2}}	Second app name	api, server
{{APP_1_DESCRIPTION}}	First app description	Web frontend
{{APP_2_DESCRIPTION}}	Second app description	API server
{{APP_1_FRAMEWORK}}	First app framework	Next.js, Vue
{{APP_2_FRAMEWORK}}	Second app framework	Express, Hono
{{APP_1_DEV_PORT}}	First app dev port	3000
{{APP_2_DEV_PORT}}	Second app dev port	8787
{{PACKAGE_NAME_1}}	First package name	ui, shared
{{PACKAGE_NAME_2}}	Second package name	utils, types
{{PACKAGE_1_DESCRIPTION}}	First package description	Shared UI components
{{PACKAGE_2_DESCRIPTION}}	Second package description	Utility functions
{{BUILD_SYSTEM}}	Monorepo build tool	Turborepo, Nx, MSBuild, Gradle, Cargo
{{MONOREPO_CONFIG}}	Config file name	turbo.json, nx.json, *.sln, settings.gradle.kts
{{APP_DIR}}	Apps directory	apps, src, services, Modules
{{PACKAGES_DIR}}	Packages directory	packages, libs, internal, pkg
{{PACKAGE_SCOPE}}	Package scope/namespace	@my-org, MyCompany, my_package
{{PROJECT_STRUCTURE}}	Full project structure	(Generated tree diagram)

Supported Ecosystems

The bootstrap command supports the following language ecosystems:

Ecosystem	Config Files	Frameworks	Monorepo Build Systems
JavaScript/TypeScript	package.json, tsconfig.json	Next.js, Nuxt, React, Vue, Angular, Express, Hono, NestJS	Turborepo, Nx, Lerna, Yarn/pnpm
PHP	composer.json, artisan	Laravel, Symfony, CodeIgniter, Slim	Laravel Modules, Symfony Flex
.NET (C#/F#)	*.csproj, *.sln	Blazor, ASP.NET Core, MAUI, WPF	MSBuild, dotnet CLI
Python	pyproject.toml, requirements.txt	Django, Flask, FastAPI, Starlette	uv, Poetry, Hatch, PDM
Ruby	Gemfile, *.gemspec	Rails, Sinatra, Hanami	Packwerk (Shopify)
Go	go.mod, go.work	Gin, Echo, Fiber, Chi	Go Workspaces
Rust	Cargo.toml	Actix, Axum, Rocket, Warp	Cargo Workspaces
Java/Kotlin	pom.xml, build.gradle.kts	Spring Boot, Quarkus, Micronaut	Gradle multi-project, Maven

File Descriptions

AGENTS.md

The primary context file for AI agents. Contains:

• Project overview and tech stack
• Directory structure
• Coding standards and patterns
• API patterns and examples
• Testing guidelines

.cursor/rules/*.mdc

Cursor IDE behavior rules:

• coding-standards.mdc - Language and framework standards
• project-context.mdc - Project documentation requirements
• version-control.mdc - Git and commit conventions
• what-to-avoid.mdc - Anti-patterns and deprecated practices
• api-server.mdc - API server patterns
• composition-api-component-structure.mdc - Vue Composition API patterns
• core-libraries.mdc - Core library usage
• environment-tooling.mdc - Environment and tooling configuration
• logging-comments.mdc - Logging and code comment conventions
• prisma-migrations.mdc - Prisma database migrations
• server-client-separation.mdc - Server/client code separation
• testing-quality.mdc - Testing standards and quality
• ui-accessibility.mdc - UI and accessibility guidelines

.cursor/commands/*.md

Custom Cursor commands for workflows:

• bootstrap.md - Start here! Analyze project and customize templates
• specify.md - Create feature specifications
• plan.md - Generate implementation plans
• implement.md - Execute implementation tasks
• review.md - Code review workflow
• commit-push.md - Git commit and push workflow
• constitution.md - Project constitution and principles
• tasks.md - Task management workflow

.github/prompts/*.prompt.md

Reusable prompts for GitHub Copilot:

• bootstrap.prompt.md - Start here! Analyze project and customize templates
• commit-push.prompt.md - Conventional commit workflow
• review.prompt.md - Code review checklist
• review-pr.prompt.md - Pull request review
• implement-feature.prompt.md - Feature implementation
• implement-fixes.prompt.md - Bug fix implementation
• specify.prompt.md - Feature specification
• plan.prompt.md - Implementation planning
• tasks.prompt.md - Task management
• assign-tasks.prompt.md - Task assignment workflow
• playwright-test.prompt.md - Playwright test generation

.github/instructions/*.instructions.md

GitHub Copilot instruction files:

• project-context.instructions.md - Architecture overview
• patterns.instructions.md - Coding patterns
• workflows.instructions.md - Development workflows
• documentation.instructions.md - Documentation standards for features and fixes

docs/* (Documentation Structure)

Documentation folders for tracking features and fixes:

• docs/features/ - Feature documentation with specs and plans
• docs/fixes/ - Fix documentation with monthly logs and complex fix folders
• docs/architecture/ - Architecture Decision Records (ADRs)
• docs/api/ - API specifications
• templates/ - Reusable documentation templates

Monorepo Note: In monorepos, app-specific docs live in apps/{app}/docs/ and root docs/ is only for monorepo-wide concerns. See monorepo documentation.instructions.md for details.

Documentation Best Practices

Feature Documentation:

• Create folder: docs/features/{{ISSUE_ID}}-FEATURE-NAME/
• Include spec.md (functional specification) and plan.md (implementation plan)
• Use /spec and /plan commands for automatic generation

Fix Documentation:

• Simple fixes: Add to monthly log docs/fixes/{YYYY-MM}.md
• Complex fixes: Create folder docs/fixes/{{ISSUE_ID}}-FIX-NAME/ with spec and plan
• Use Keep a Changelog format

Decision Tree: Is fix complex (multi-file, architectural)? → Folder. Otherwise → Monthly log.

See documentation.instructions.md for comprehensive guidelines.

AI Agent Skills

Enhance your AI agents with specialized capabilities using the Agent Skills open format. Skills are folders of instructions, scripts, and resources that agents can discover and use to perform tasks more accurately and efficiently.

Skills Resources

Resource	Description
agentskills.io	Official specification, documentation, and integration guides for the Agent Skills format
skills.sh	Community skills directory with leaderboard and one-command installation

How Skills Work

Skills use progressive disclosure to manage context efficiently:

1. Discovery: Agents load only the name and description of each skill
2. Activation: When a task matches a skill's description, the agent reads full instructions
3. Execution: The agent follows instructions, loading referenced files or code as needed

Skill Structure

At its core, a skill is a folder containing a SKILL.md file:

my-skill/
├── SKILL.md        # Required: instructions + metadata
├── scripts/        # Optional: executable code
├── references/     # Optional: documentation
└── assets/         # Optional: templates, resources

Installing Skills

Install skills from skills.sh using the -a or --agent flag to specify your AI agent:

# For a single agent
npx -y skills add -a cursor <owner/repo>

# For multiple agents (use multiple -a flags)
npx -y skills add -a cursor -a copilot <owner/repo>

Agent Detection

The skills CLI automatically detects which agents you have installed when you run any command. If you don't specify the -a flag, you'll be prompted to select from your detected agents.

Supported agents include:

• Cursor, GitHub Copilot, Windsurf, Cline, Continue, Aider, Zed, Roo Code, and 35+ more
• See full list of supported agents

Agent-Specific Installation Examples

# For GitHub Copilot only
npx -y skills add -a copilot <owner/repo>

# For Cursor only
npx -y skills add -a cursor <owner/repo>

# For Windsurf only
npx -y skills add -a windsurf <owner/repo>

# For Cline only
npx -y skills add -a cline <owner/repo>

# For both Cursor and GitHub Copilot
npx -y skills add -a cursor -a copilot <owner/repo>

Important: Always specify the agent(s) explicitly using -a <agent-name> to install only for the agents you actively use. You can specify multiple agents using multiple -a flags. This prevents creating unnecessary configurations for unused agents.

Skills are installed to {.cursor|.agents|.github}/skills/ and automatically available to your selected AI agent(s).

Bundled Universal Skills

This toolkit includes 16 pre-installed universal skills in the .agents/skills/ folder. These skills are copied automatically when you set up your project and require no additional installation.

Skill	Description
test-driven-development	TDD practices with red-green-refactor workflow. Write failing test first, then minimal code to pass.
systematic-debugging	Structured debugging methodology. Find root cause before attempting fixes—no guessing.
verification-before-completion	Quality checks before claiming work is done. Evidence before assertions, always.
writing-plans	Creating clear implementation plans before coding.
executing-plans	Following through on implementation plans systematically.
requesting-code-review	Best practices for requesting code reviews from teammates.
receiving-code-review	Responding constructively to review feedback.
brainstorming	Structured ideation and problem-solving sessions.
writing-skills	Creating effective SKILL.md files for custom skills.
using-superpowers	Meta-skill for leveraging the full skill system.
using-git-worktrees	Git worktree workflows for parallel development.
dispatching-parallel-agents	Coordinating multiple AI agents on related tasks.
subagent-driven-development	Breaking complex tasks into subagent-manageable chunks.
finishing-a-development-branch	Completing and merging development branches cleanly.
logging	Structured logging standards, log levels, and observability patterns.
project-documentation	README files, code comments, ADRs, and changelog best practices.

These skills are ready to use immediately after copying the templates. No npx -y skills add -a <agent> commands needed.

Additional Recommended Skills

Beyond the bundled skills, these additional skills complement this toolkit for specific use cases:

Note: All install commands in the tables below should include the -a <agent> flag to specify your AI agent. For example:

• npx -y skills add -a copilot obra/superpowers (for GitHub Copilot only)
• npx -y skills add -a cursor obra/superpowers (for Cursor only)
• npx -y skills add -a cursor -a copilot obra/superpowers (for both Cursor and GitHub Copilot)

Tip: If you omit the -a flag, the CLI will automatically detect your installed agents and prompt you to choose.

Development Workflow Skills (External)

If you want the original obra/superpowers skills (which the bundled skills are derived from), or additional workflow skills:

Skill	Install Command	Description
Full Superpowers Suite	npx -y skills add obra/superpowers	All workflow skills from the original source

Security & Quality Skills

Skill	Install Command	Description
Semgrep	npx -y skills add trailofbits/skills	Static analysis patterns
Property-Based Testing	npx -y skills add trailofbits/skills	Testing with property-based approaches
Code Review	npx -y skills add skillcreatorai/Ai-Agent-Skills	Comprehensive code review guidance
Logging Best Practices	npx -y skills add boristane/agent-skills	Structured logging patterns

Frontend Skills

Skill	Install Command	Description
Vercel React Best Practices	npx -y skills add vercel-labs/agent-skills	React patterns and optimization
Web Design Guidelines	npx -y skills add vercel-labs/agent-skills	UI/UX design principles
Frontend Design	npx -y skills add anthropics/skills	General frontend patterns
TailwindCSS Setup	npx -y skills add expo/skills	Tailwind configuration

Framework-Specific Skills

Skill	Install Command	Description
Vue	npx -y skills add onmax/nuxt-skills	Vue.js best practices
Nuxt	npx -y skills add onmax/nuxt-skills	Nuxt framework patterns
Nuxt UI	npx -y skills add onmax/nuxt-skills	Nuxt UI component library
Better Auth	npx -y skills add better-auth/skills	Authentication patterns
Expo	npx -y skills add expo/skills	React Native with Expo
NestJS	npx -y skills add Kadajett/agent-nestjs-skills	NestJS backend patterns
Elysia.js	npx -y skills add elysiajs/skills	Elysia.js API patterns
Three.js	npx -y skills add CloudAI-X/threejs-skills	Three.js 3D graphics
Remotion	npx -y skills add remotion-dev/skills	Remotion video creation
Convex	npx -y skills add waynesutton/convexskills	Convex backend patterns
TanStack	npx -y skills add jezweb/claude-skills	TanStack Query patterns
SwiftUI	npx -y skills add Dimillian/Skills	SwiftUI iOS patterns
Obsidian	npx -y skills add kepano/obsidian-skills	Obsidian note-taking
shadcn/ui	npx -y skills add giuseppe-trisciuoglio/developer-kit	shadcn/ui components
Stripe	npx -y skills add anthropics/claude-plugins-official	Stripe integration

Language-Specific Skills

Language/Framework	Install Command	Included Skills
PHP	npx -y skills add vapvarun/claude-backup --skill "php"	PHP best practices
Laravel	npx -y skills add vapvarun/claude-backup --skill "laravel"	Laravel patterns
Python	npx -y skills add siviter-xyz/dot-agent --skill "python"	Python best practices
Django	npx -y skills add vintasoftware/django-ai-plugins --skill "django-expert"	Django patterns
Next.js	npx -y skills add sickn33/antigravity-awesome-skills --skill "nextjs-best-practices"	Next.js best practices
React	npx -y skills add vercel-labs/agent-skills --skill "vercel-react-best-practices"	React + Vercel patterns
Vue	npx -y skills add onmax/nuxt-skills --skill "vue"	Vue.js best practices
Nuxt	npx -y skills add onmax/nuxt-skills --skill "nuxt"	Nuxt patterns
Expo	npx -y skills add expo/skills	Expo/React Native
TypeScript	npx -y skills add pproenca/dot-skills	TypeScript best practices
Advanced Types	npx -y skills add wshobson/agents	TypeScript advanced type patterns

Note: For frameworks without dedicated skills, use npx -y skills add -a <agent> anthropics/skills which includes skill-creator to help you create custom skills. Remember to replace <agent> with your AI agent name (e.g., copilot, cursor, windsurf, cline).

Documentation & Writing Skills

Skill	Install Command	Description
Crafting Effective READMEs	npx -y skills add softaworks/agent-toolkit	README writing guidance
Writing Clearly and Concisely	npx -y skills add softaworks/agent-toolkit	Clear technical writing
Doc Coauthoring	npx -y skills add anthropics/skills	Collaborative documentation

Adding Skills to Your Project

After bootstrapping your project with this toolkit:

1. Identify your tech stack - Review your framework and tooling
2. Install relevant skills - Add skills that match your technologies using -a <agent> flags
3. Add custom skills - Create project-specific skills as needed

Example for a Nuxt + Hono monorepo:

# Note: Development workflow skills (TDD, debugging, planning, etc.)
# are already bundled in .agents/skills/ - no installation needed!

# For Cursor only
npx -y skills add -a cursor onmax/nuxt-skills
npx -y skills add -a cursor vercel-labs/agent-skills
npx -y skills add -a cursor better-auth/skills
npx -y skills add -a cursor trailofbits/skills

# Or, for multiple agents (e.g., Cursor + GitHub Copilot)
npx -y skills add -a cursor -a copilot onmax/nuxt-skills
npx -y skills add -a cursor -a copilot vercel-labs/agent-skills
npx -y skills add -a cursor -a copilot better-auth/skills
npx -y skills add -a cursor -a copilot trailofbits/skills

Creating Custom Skills

You can create project-specific skills following:

• Official Specification - Complete format specification for SKILL.md files
• Authoring Best Practices - Guidelines for writing effective skills
• Skill Creator on skills.sh - Community examples and templates

A minimal SKILL.md requires only name and description in YAML frontmatter:

---
name: my-custom-skill
description: When to use this skill and what it does.
---

# My Custom Skill

## When to use this skill

Use this skill when the user needs to...

## Instructions

1. Step one...
2. Step two...

Custom skills are useful for:

• Organization-specific coding standards
• Internal library documentation
• Domain-specific patterns
• Team workflow procedures

Skill Directory Structure

After copying templates and installing additional skills, your project will have:

.agents/                  # Bundled universal skills (from this toolkit)
└── skills/
    ├── test-driven-development/
    │   └── SKILL.md
    ├── systematic-debugging/
    │   └── SKILL.md
    ├── verification-before-completion/
    │   └── SKILL.md
    └── ... (14 skills total)

.cursor/                  # or .github/ for GitHub Copilot
├── rules/                # IDE behavior rules (from this toolkit)
├── commands/             # Custom commands (from this toolkit)
└── skills/               # Additional skills from skills.sh
    ├── nuxt-skills/      # From onmax/nuxt-skills
    │   └── SKILL.md
    ├── agent-skills/     # From vercel-labs/agent-skills
    │   └── SKILL.md
    └── {skill-name}/     # Each installed skill gets its own folder
        └── SKILL.md

Skill Locations:

Location	Contents	Source
.agents/skills/	Universal workflow skills (TDD, debugging, planning, etc.)	Bundled with this toolkit
.cursor/skills/	Framework/language-specific skills	Installed via npx -y skills add -a <agent>
.github/skills/	Alternative location for GitHub Copilot	Installed via npx -y skills add -a <agent>

Note: Skills installed via npx -y skills add -a <agent> use the skill name as the folder name (not org/repo path). For example, npx -y skills add -a cursor obra/superpowers installs to .cursor/skills/superpowers/.

Recommendation: Only install skills for agents you have installed and actively use. Symlinking to all known agents creates unnecessary configuration files and can clutter your project.

Note: Skills complement the templates in this toolkit. The templates provide project structure and conventions, while skills provide procedural knowledge for specific technologies.

Contributing

1. Fork this repository
2. Create a feature branch
3. Make your changes
4. Submit a pull request

License

MIT License - See LICENSE for details.