Skip to main content
Glama
development-guide.mdc1.9 kB
--- description: globs: alwaysApply: false --- # Development Guide This guide covers the development workflow for the Rakit UI AI project. ## Project Setup - [package.json](mdc:package.json) defines the project dependencies and scripts - TypeScript configuration is in [tsconfig.json](mdc:tsconfig.json) - The project uses ES modules (type: "module" in package.json) - Check [TODO.md](mdc:TODO.md) for planned improvements and features ## Build & Run - `npm run build`: Cleans dist directory, compiles TypeScript, and builds MCP - `npm run watch`: Watches for changes during development - `npm start`: Starts the MCP server from the compiled dist directory ## Key Dependencies - `@modelcontextprotocol/sdk`: Core MCP SDK - `mcp-framework`: Framework for building MCP tools - `open`: Library for opening URLs in the default browser ## Development Workflow 1. Create/modify tool implementations in `src/tools/` 2. Update templates as needed in `src/tools/templates/` 3. Run `npm run watch` for development 4. Test using `npm start` 5. Build for distribution with `npm run build` ## Working on Improvements 1. Check [TODO.md](mdc:TODO.md) for prioritized tasks 2. Create feature branches for each improvement 3. Follow the security guidelines in security-best-practices rule 4. Add tests according to testing-strategy rule 5. Optimize performance following performance-optimization rule ## Testing - Unit tests: `npm test` (once implemented) - Integration tests: `npm run test:integration` - E2E tests: `npm run test:e2e` - Coverage report: `npm run test:coverage` ## Contributing 1. Pick a task from TODO.md 2. Create a feature branch 3. Implement with tests 4. Update documentation 5. Submit pull request ## Note on Browser Integration The server creates temporary files in a `tmp` directory at the project root. Make sure this directory has appropriate permissions for file creation.

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/madebyaris/rakitui-ai'

If you have feedback or need assistance with the MCP directory API, please join our Discord server