agent-native  by BuilderIO

Build agent-native applications with deeply integrated UIs and autonomous agents

Created 2 months ago
382 stars

Top 74.7% on SourcePulse

GitHubView on GitHub
1 Expert Loves This Project
Project Summary

Agent-Native Applications Framework

This framework addresses the challenge of building applications that seamlessly blend structured user interfaces with autonomous AI agents. It targets developers and power users seeking to create owned, customizable, and self-improving applications, offering a unified system where agents and UIs are equal citizens, synchronizing state and actions instantly. The primary benefit is achieving the polish of SaaS tools with the deep flexibility and ownership of raw AI agents.

How It Works

Agent-Native treats agents and UIs as first-class citizens within a single system, enabling bidirectional actions and shared state synchronization via a unified database. Agents are context-aware, leveraging user selections or interactions to perform tasks. The architecture supports per-user workspaces, managing skills, memory, and sub-agents with SQL backing. It facilitates agent-to-agent communication (A2A) for complex workflows and allows agents to autonomously enhance the application over time by adding features or refining the UI. The system is designed for flexibility, supporting any SQL database compatible with Drizzle and any hosting target supported by Nitro, while allowing the use of diverse AI models.

Quick Start & Requirements

  • Primary install/run command: npx @agent-native/core create my-platform followed by cd my-platform, pnpm install, and pnpm dev.
  • Non-default prerequisites: pnpm package manager, and environment variables for API keys (e.g., ANTHROPIC_API_KEY, BUILDER_PRIVATE_KEY, A2A_SECRET).
  • Links: Full documentation available at agent-native.com.

Highlighted Details

  • Cloneable SaaS Templates: Offers production-grade, forkable SaaS application templates (e.g., Mail, Calendar, Content, Slides, Video, Analytics, Design, Forms) that replace existing tools, providing full ownership.
  • Self-Improving Applications: Agents can autonomously add features, fix bugs, and refine the UI, leading to continuously evolving applications.
  • Unified Multi-App Workspaces: Supports monorepo structures for multiple applications sharing authentication and enabling zero-config cross-app agent communication.
  • Flexible AI and Data Stack: Compatible with any SQL database supported by Drizzle, any hosting target supported by Nitro, and a variety of AI agents including Claude Code, Codex, Gemini CLI, OpenCode, and Builder.io.

Maintenance & Community

  • Community: A Discord server is available for questions, sharing, and assistance.
  • Documentation: Comprehensive documentation is provided at agent-native.com.

Licensing & Compatibility

  • License Type: MIT.
  • Compatibility: The MIT license permits commercial use and integration within closed-source projects without significant restrictions.

Limitations & Caveats

No explicit limitations such as alpha status, known bugs, or unsupported platforms were detailed in the provided README. The architecture implies a dependency on specific AI providers and requires careful management of API keys and environment variables.

Health Check
Last Commit

23 hours ago

Responsiveness

Inactive

Pull Requests (30d)
531
Issues (30d)
3
Star History
318 stars in the last 30 days

Explore Similar Projects

Starred by Elie Bursztein Elie Bursztein(Cybersecurity Lead at Google DeepMind), Chip Huyen Chip Huyen(Author of "AI Engineering", "Designing Machine Learning Systems"), and
7 more.

SuperAGI by TransformerOptimus

0.1%
18k
Open-source framework for autonomous AI agent development
Created 3 years ago
Updated 1 year ago
Feedback? Help us improve.