Vibesdk

vibesdk is an agentic software development platform and framework designed to coordinate autonomous agents that write, debug, and refine full-stack applications from natural language. It serves as a cloud-native application orchestrator and an LLM-powered code generation framework that converts prompts into functional code through iterative conversations and multi-phase agent behaviors.

The project distinguishes itself by providing a complete toolchain for building AI development platforms. This includes the ability to integrate various model providers, construct custom LLM toolkits, and manage the entire lifecycle of AI-generated apps via a serverless deployment toolchain and a programmatic TypeScript SDK.

The platform covers a broad range of capabilities, including AI sandbox orchestration for isolated execution and live previews, Git-backed virtual filesystems for version tracking, and automated cloud deployment to serverless worker platforms. It also incorporates systems for database schema management, hierarchical secret encryption, and real-time state synchronization via WebSockets.

Users can manage project workflows through a command line interface or programmatically using the provided SDK.

Features

AI Web Application Builders - Generates full-stack web applications from natural language product requirements using coordinated AI agents.
Software Development Agents - Provides a framework for coordinating autonomous agents to automate full-stack software engineering, debugging, and refinement workflows.
Full-Stack Code Generators - Generates complete full-stack web applications, including both frontend and backend code, from natural language conversations.
Custom Tool Registrations - Allows the creation and registration of custom toolkits to expand the functional capabilities of integrated language models.
Phase-Based Generation Pipelines - Coordinates the application generation process through a structured timeline of distinct stages from prompting to deployment.
AI Execution Sandboxes - Orchestrates isolated containerized environments specifically designed for running AI coding agents securely.
AI Provider Managers - Manages API keys and endpoint settings for connecting to various hosted or custom AI model providers.
Iterative Generative Coding - Implements a conversational loop where LLMs refine features and fix bugs through iterative code modifications.
Application Generation - Creates new applications or components using natural language prompts and configurable generation behaviors.
Conversational Layout Refinement - Enables iterative modification of generated UI code through conversational feedback loops using text and image messages.
AI Development Platforms - Provides a framework for building custom AI development platforms by integrating model providers and toolkits.
Code Refinement - Implements a conversational interface for iterating on features and fixing errors through AI-driven code refinement.
Git Virtual File Systems - Tracks code versions and project snapshots using a database-driven filesystem compatible with Git.
Stateful Durable Objects - Utilizes distributed globally consistent objects to store application metadata and session state for real-time synchronization.
Platform SDKs - Ships a programmatic TypeScript SDK for building AI-driven application generation and deployment platforms.
Containerized Sandbox Runtimes - Runs generated application code in isolated containerized environments for secure previews and testing.
Application Cloud Deployments - Pushes built application bundles to cloud platforms for live hosting and production access.
Cloud Native Orchestration - Manages the lifecycle of AI-generated applications, including sandboxed execution and cloud-native deployment.
Serverless Deployment - Pushes generated application bundles to a cloud-native serverless runtime for instant global hosting.
Application Isolation Containers - Runs generated applications in isolated containerized environments to ensure security and prevent interference.
Code Sandboxing Environments - Sets up containerized workspaces with build tools to isolate and execute generated application code.
Live Preview Sandboxes - Runs generated code in sandboxed containers to provide immediate visual feedback and execution testing.
AI-Powered Code Fixes - Provides automated code correction by analyzing runtime errors and logs using reasoning models to fix multiple files in parallel.
Full-Stack Cloud Deployments - Automates the deployment of full-stack applications, including frontend and backend, to a serverless cloud stack.
WebSocket State Synchronization - Streams application state and model responses to the client in real time using WebSockets for immediate feedback.
Model Streaming Over WebSocket - Streams model-generated responses and build progress updates to the client via persistent WebSocket connections.
Application State Management - Implements systems for managing application-wide state using durable objects and relational databases.
Database Schema Managers - Provides tools for defining, versioning, and migrating database structures across local and production environments.
GitHub Integrations - Transfers generated project files directly into GitHub repositories for further version-controlled development.
Session Restoration - Allows recovery of project state and generation timelines by reconnecting to existing agent sessions.
Git Workspace Isolations - Uses a Git-backed filesystem to provide isolated workspaces for deploying and previewing applications.
Application Generation APIs - Provides a TypeScript SDK to programmatically control the application generation and deployment process.
Encrypted Secret Management - Protects API keys and credentials using a hierarchical derivation system for secure access and rotation.
API Key Encryption at Rest - Encrypts user secrets using a hierarchical key derivation system with support for rotation and expiration.
Secrets Management Platforms - Protects the platform through secret encryption, rate limiting, and input validation.
Virtual Filesystem Versioning - Tracks code changes using a Git-compatible system with commit history stored in a persistent database.
Task Progress Monitors - Tracks the real-time status of the application build process through a phase-based timeline and progress updates.
API Endpoint Definitions - Defines types and controllers to expose backend functionality through a structured routing system.
Preview Deployments - Generates live, shareable preview environments for specific application branches directly from the backend.
Command Line Tools - SDK specifically designed for building vibe coding applications.

serverless-stack/serverless-stack

26,112View on GitHub

This project is a serverless full-stack framework and infrastructure-as-code tool designed for deploying and managing full-stack applications. It functions as a cloud-native deployment orchestrator that synchronizes application code and environment variables across frontend content delivery networks and backend API endpoints. The platform enables the deployment of complete application architectures on either private or cloud infrastructure. It allows for the orchestration of self-hosted applications without the need to manage traditional virtual servers. The system covers serverless infrastr

cloudflare/agents

3,466View on GitHub

This is an open-source framework for building stateful, durable AI agents that run on Cloudflare Workers. It provides a runtime for long-lived agents that maintain a persistent identity, local SQL storage, and real-time connections, utilizing a lifecycle where agents hibernate when idle and wake on demand. The project distinguishes itself through its multi-channel orchestration, allowing a single agent to be deployed across voice, email, and chat interfaces with unified state. It implements the Model Context Protocol for standardized tool and data exchange and includes a dedicated framework f

tldraw/draw-a-ui

5,445View on GitHub

This project is an AI-powered visual canvas and collaborative whiteboard framework. It functions as a customizable vector drawing engine and a tool for converting hand-drawn interface sketches and wireframes into functional code using artificial intelligence. The system distinguishes itself through the integration of AI agents that can read, modify, and generate visual diagrams directly on the canvas. It also provides a node-based workflow editor for building automation pipelines and data processing flows by connecting multimodal components. The platform covers a broad range of capabilities,

rivet-dev/rivet

5,619View on GitHub

Rivet is a distributed infrastructure for managing the lifecycle, addressing, and persistence of stateful actors and durable execution engines. It provides a distributed process sandbox that executes application logic within lightweight isolates, ensuring resource isolation and fast cold starts. The system is designed to coordinate multi-step operations using persistent queues and timers to guarantee reliable task completion across distributed environments. The platform specifically enables the orchestration of stateful AI agents that maintain persistent memory and state across long-running i

cloudflarevibesdk

Features