React Native developers often struggle to find reusable UI components and animation examples scattered across various platforms. rnui.dev addresses this by providing a centralized, searchable catalog of curated React Native UI components, animations, and design inspiration. It leverages AI, specifically OpenAI Codex, to assist in the contribution and review process, aiming to save developers significant time and effort in sourcing references.

How It Works

The project integrates OpenAI Codex (gpt-4o-mini and text-embedding-3-small) into its core workflow. For contributions, a codex:ingest script fetches repository metadata, uses Codex to generate structured catalog entries, validates them, appends them to the appropriate data files, and automatically opens a pull request. A GitHub Actions workflow (codex-triage.yml) employs Codex to score incoming PRs against a defined rubric, posting reviews directly as comments. For search, codex:index generates embeddings for catalog entries, stored locally in data/embeddings.json, enabling natural-language search queries via cosine similarity through the /api/search endpoint.

Quick Start & Requirements

• Installation: Clone the repository (git clone https://github.com/mrpmohiburrahman/rnui.dev), navigate into the directory (cd rnui.dev), and install dependencies using pnpm install.
• Prerequisites: Requires pnpm for package management. An OPENAI_API_KEY is essential for AI-assisted ingestion, review, and indexing. Additional keys for Supabase, Firebase, and ImageKit are needed in the .env.local file (copied from .env.example).
• Running: Execute pnpm dev to start the local development server. Access the application at http://localhost:3000.
• Contribution: The fastest contribution method is opening an issue with the "Animation Submission" template and providing the source URL. Detailed contribution flows are available in CONTRIBUTING.md.
• Live Demo: https://rnui.dev

Highlighted Details

• AI-assisted ingestion and review powered by OpenAI Codex.
• Automated pull request scoring and feedback mechanism.
• Natural-language search functionality for UI components and animations.
• Catalog organized into specific UI-element categories (e.g., accordions, buttons, charts, sliders).

Maintenance & Community

The repository is maintained by mrpmohiburrahman. No specific community channels (like Discord/Slack) or details on major contributors/sponsorships are provided in the README.

Licensing & Compatibility

The project is licensed under the MIT license. This license is permissive and generally compatible with commercial use and integration into closed-source projects.

Limitations & Caveats

All AI-powered features require a valid OPENAI_API_KEY and will fail if it is missing. Demo video uploads are currently a manual process. The search index relies on a local JSON file (data/embeddings.json), with plans to migrate to Supabase pgvector as the catalog scales beyond JSON's capabilities. The AI review workflow is gated by the OPENAI_API_KEY secret.