Coding & Development🤖claude⌨️cursor🌊windsurf🌐universal
10xRules.ai by 10xDevs
Generate "Rules for AI". Quickly ✨
Quick Start
Install this Skill
npx skills add przeprogramowani/ai-rules-builder# 10xRules.ai by 10xDevs
<!-- ALL-CONTRIBUTORS-BADGE:START - Do not remove or modify this section -->
[](#contributors-)
<!-- ALL-CONTRIBUTORS-BADGE:END -->
[](https://github.com/przeprogramowani/ai-rules-builder/actions/workflows/deploy-app.yml)

👉 [10xRules.ai](https://10xrules.ai)
Create so called "rules for AI" written in Markdown, used by tools such as GitHub Copilot, Cursor and Windsurf, through an interactive, visual interface.
## Features
- **Build AI Rules:** Create customized rule sets for different editors (Copilot, Cursor, Windsurf)
- **Export Options:** Easily copy to clipboard or download as markdown files
- **Smart Import:** Automatically generate rules by dropping package.json or requirements.txt files
- **Editor Integration:** Provides programmatic access to rules via an [MCP Server](./mcp-server/README.md) for integration with AI assistants in editors like Cursor
## Getting Started
1. **Installation**
```bash
npm install
```
2. **Development**
```bash
npm run dev
```
3. **Build**
```bash
npm run build
```
## Local Supabase Setup
1. Install Supabase CLI
Check [Supabase docs](https://supabase.com/docs/guides/local-development?queryGroups=package-manager&package-manager=brew#quickstart) for local setup reference.
```bash
brew install supabase/tap/supabase
```
2. Start Supabase (requires local Docker)
```bash
supabase start
```
3. Verify Supabase is running
```bash
supabase status
```
## Dotenv
Create `.env.local` file and add the following variables:
```bash
PUBLIC_ENV_NAME=local
SUPABASE_URL=http://localhost:54321
SUPABASE_PUBLIC_KEY=your_supabase_anon_key
SUPABASE_SERVICE_ROLE_KEY=your_supabase_service_role_key
CF_CAPTCHA_SITE_KEY=1x00000000000000000000AA
CF_CAPTCHA_SECRET_KEY=1x0000000000000000000000000000000AA
# Email delivery via Resend (used by Supabase SMTP)
# Get your API key from https://resend.com
RESEND_API_KEY=re_xxxxxxxxxxxxx
```
## Tech Stack
- Astro 5
- TypeScript 5
- React 18.3
- Tailwind 4
- Zustand
- Lucide React
## Project Components
This repository contains multiple key components:
- **AI Rules Builder UI (Root):** The main Astro/React application providing the web interface for creating and managing AI rules.
- **MCP Server (`./mcp-server`):** A Cloudflare Worker implementing the Model Context Protocol (MCP). This server allows AI assistants (like Cursor, Claude, etc.) to programmatically access the defined AI rules via specific tools (`listAvailableRules`, `getRuleContent`). This enables integration with editors for fetching context-aware coding guidelines. For detailed setup, usage, and planned features, see the [MCP Server README](./mcp-server/README.md).
### Feature Flags
The project uses a feature flags system to separate deployments from releases. Feature flags can be used to control functionality availability based on the environment (`local`, `integration`, `prod`). The system supports flags for:
- API endpoints
- Astro pages
- UI components visibility
For detailed documentation about feature flags implementation, see `.ai/feature-flags.md`.
### Testing
This project uses a comprehensive testing stack including unit tests and end-to-end tests.
### Unit Testing with Vitest
Unit tests are implemented using Vitest with JSDOM for browser environment simulation and React Testing Library for component testing.
Available commands:
```bash
# Run unit tests
npm run test
# Run tests in watch mode
npm run test:watch
# Run tests with UI
npm run test:ui
# Generate coverage report
npm run test:coverage
```
### End-to-End Testing with Playwright
E2E tests are implemented using Playwright with the Page Object Model pattern for maintainable tests.
Available commands:
```bash
# Run E2E tests
npm run test:e2e
# Run E2E tests with UI
npm run test:e2e:ui
# Generate test code with codegen
npm run test:e2e:codegen
```
### Test Structure
- `tests/unit/` - Unit tests
- `tests/setup/` - Test setup files
- `e2e/` - End-to-end tests
- `e2e/page-objects/` - Page Object Model classes
- `e2e/fixtures/` - Test fixtures and data
### CI/CD Integration
Tests are automatically run in the CI/CD pipeline using GitHub Actions. See `.github/workflows/tests.yml` for configuration.
## Contributions
Send updates to:
- `src/data/dictionaries.ts`
- `src/data/rules/...`
Important: Introduce translations for new rules in `src/i18n/translations.ts`, otherwise the unit test will fail.
## How to Write Effective Rules
When contributing new rules, please:
- **Be specific:** "Use React.memo for expensive components" not "Optimize components"
- **Make it actionable:** Provide clear guidance that can be immediately applied
- **Include placeholders:** Use `{{placeholder_te
... (truncated)Repository
234stars
56forks
Details
Related Skills
More coding & development skills you might like
Reticle
Coding & DevelopmentAI agents can generate code, but still struggle to understand what they build. Reticle brings Jev-style machine-native runtime perception to web & desktop applications.
🤖claude🌐universal
1.2k
190
View
agentscope-java
Coding & DevelopmentExpert Java developer skill for AgentScope Java framework - a reactive, message-driven multi-agent system built on Project Reactor. Use when working with reactive programming, LLM integration, agent orchestration, multi-agent systems, or when the user mentions AgentScope, ReActAgent, Mono/Flux, Project Reactor, or Java agent development. Specializes in non-blocking code, tool integration, hooks, pipelines, and production-ready agent applications.
🌐universal
6.0k
1.5k
View
ospec
Coding & DevelopmentDocument-driven OSpec workflow for initialization, change/goal routing, validation, archiving, and durable project knowledge.
✨gemini🤖claude💬gpt🌐universal
524
34
View
misakanet-failure-memory
Coding & Development📚 A zero-dependency, git-backed micro-lesson library for AI Agents to asynchronously share and search verified debugging experience. Python stdlib only. | https://misakanet.org
🤖claude⌨️cursor🌐universal
424
158
View