
Vibekit

VibeKit is a hosted AI coding platform that gives each web app its own persistent AI agent and development environment. Users can build, edit, deploy, and maintain web apps from a phone or browser without keeping their own computer online. VibeKit supports GitHub integration, scheduled tasks, databases, MCP, environment variables, custom domains, and bring your own API keys for supported AI providers. The platform is designed for developers and founders who want to build small web apps, PWAs, prototypes, or internal tools with an AI coding agent that can continue working across sessions.
Vibekit Details
Reviewed by Add AI Directory Editorial Team
View our review methodology →Sources
- Official website
- DocumentationNot provided
- Pricing pageNot provided
- Privacy policyNot provided
- Terms / public policyNot provided
- Testing status
- Researched Only
- Availability status
- Not yet verified
- Limitations
- No additional limitations documented.
Ready to try Vibekit?
Check out Vibekit for pricing and explore how it can streamline your workflow.
VibeKit is a hosted AI coding agent platform that gives each web app its own persistent AI agent and isolated development environment. The agent can edit code, inspect logs, run tasks, and deploy changes while VibeKit keeps the application running on its hosted infrastructure.
What Is VibeKit?
VibeKit allows developers and founders to build and maintain web applications without keeping their own computer online. Each project receives a dedicated container and persistent AI agent that retains access to the application's codebase between sessions.
Users can interact with the agent through VibeKit's iOS app, web dashboard, CLI, REST API, or MCP. A user can ask the agent to modify an application, fix a bug, add a feature, or deploy a change from a phone or browser.
VibeKit can create or connect to a GitHub repository, commit changes to the repository, and automatically redeploy an application after changes are pushed. The GitHub repository remains under the user's account and serves as the source of truth for the project's code.
Key Features
Persistent AI Agent Per App
VibeKit assigns a separate AI coding agent to each application. The agent works inside that application's environment and maintains project context across sessions instead of operating as one shared assistant across unrelated projects.
Hosted Development Environment
VibeKit runs application containers on its own infrastructure. Users do not need to keep a laptop, desktop, or home server running for the agent or deployed application to remain available.
Build and Deploy From a Phone
VibeKit provides an iOS application alongside its web interface. Developers can send coding instructions, review a project, and initiate changes without working directly from a traditional development machine.
GitHub Integration
VibeKit can work with a GitHub repository owned by the user. The agent can edit files, create commits, work with branches, and deploy changes after pushes to the repository. Existing GitHub repositories can also be connected to the platform.
MCP, CLI and API Access
Developers can interact with VibeKit through MCP, a command line interface, and a REST API in addition to the web and mobile interfaces. This allows VibeKit agents to participate in broader AI agent and development workflows.
Scheduled Agent Tasks
VibeKit agents can perform work on a schedule in addition to responding to direct requests. This makes the platform suitable for applications that require recurring maintenance or automated development tasks.
Bring Your Own AI Key
VibeKit supports bring your own key access for supported Anthropic and OpenAI credentials. With BYOK, AI usage is billed directly through the provider without an additional VibeKit markup. Users who do not want to supply their own credentials can use VibeKit's platform billing instead.
VibeKit Use Cases
VibeKit can be used to:
- Build and deploy small web applications from a phone or browser
- Create internal business tools with an AI coding agent
- Maintain an application with a persistent agent between work sessions
- Run an AI coding agent without maintaining an always-on computer
- Connect an existing GitHub project to a hosted AI development environment
- Automate recurring development and maintenance tasks
- Build prototypes and PWAs
- Deploy applications directly to hosted URLs
How VibeKit Works
A user creates an application or connects an existing GitHub repository. VibeKit provisions a hosted container for the project and assigns a persistent AI agent to that environment.
The user then gives the agent instructions through chat or one of VibeKit's developer interfaces. The agent can inspect the existing project, edit files, run commands, test changes, and deploy the updated application.
Applications can run on a vibekit.bot subdomain, while paid plans support additional deployment options such as custom domains.
VibeKit Pricing
VibeKit uses a Free + Paid pricing model.
The Free plan starts at $0 per month. VibeKit currently lists its Builder plan at $19.99 per month and its Pro plan at $49.99 per month.
AI usage can work differently from the platform subscription. Users who bring supported provider credentials can pay the AI provider directly with no VibeKit markup. When VibeKit provides paid model access through its platform billing system, non-free model usage is charged at the underlying provider cost plus a platform markup.
Limitations
VibeKit runs source code on hosted infrastructure rather than exclusively on a user's local computer. Developers whose security requirements prohibit source code from leaving a local machine may need a locally operated alternative.
The platform focuses on web applications and hosted development workflows. It should not be confused with a general-purpose Linux environment or a tool for producing native App Store application binaries.
Free usage is also subject to platform limits, so developers building or maintaining multiple production applications may require a paid plan.
VibeKit's zero-markup AI pricing specifically applies to supported bring-your-own-key usage. Platform-provided paid AI access uses a separate billing model.
Who Is VibeKit For?
VibeKit is designed for developers, founders, technical teams, and independent builders who want an AI coding agent that remains attached to a project and runs in a hosted environment.
It is especially useful for users who want to build or maintain web applications from a phone, keep an AI development agent available between sessions, or avoid maintaining a dedicated computer for long-running AI coding workflows.
Category
Agent Tools
Pricing
Free + Paid
Free plan available. Paid plans start at $19.99/month.
Availability
Public
VibeKit is available through its web interface and iOS application.
Ready to try Vibekit?
Check out Vibekit for pricing and explore how it can streamline your workflow.
Explore More AI Agents
Discover other AI agents and tools to enhance your workflow and productivity.
Browse All AgentsSimilar to Vibekit

Screenpipe
Screenpipe is a source-available desktop app that records and indexes screen activity, app text, meetings, and audio to create a searchable history of your work. Captured data stays on your device by default and can be selectively shared with Claude, Codex, Cursor, ChatGPT, and other MCP-compatible AI assistants. Screenpipe also supports recurring workflows based on captured activity, local transcription, accessibility text extraction, and OCR. Optional paid plans add features such as extended history, additional AI capacity, cross-device syncing, and team workflows.

Agentskills Codes
agentskills.codes is an open registry for discovering and installing reusable Agent Skills for AI coding assistants such as Claude Code, Codex, and GitHub Copilot. It helps developers compare skills, review source and compatibility information, check safety and repository health signals, and install portable skills for coding and agent workflows.

Mem0
Mem0 is an AI memory layer for applications and agents that enables persistent, personalized context across conversations and sessions. It helps developers build AI systems that can remember user preferences, previous interactions, decisions, and other relevant information over time. Mem0 provides APIs and SDKs for storing, retrieving, updating, and managing memories, making it useful for AI assistants, customer support systems, personalized agents, and other context-aware AI applications.
Trending AI Agents

ListingBott
ListingBott is an AI-powered SEO tool that automates directory submissions to 100+ sites, helping you earn quality backlinks, boost domain authority, and increase online visibility—all with one click.

OpenAI Agent Builder
OpenAI Agent Builder is a visual platform for building, deploying, and managing autonomous AI agents. It provides a node-based workflow canvas, integrated tools, traceability, and safety controls that connect language models to real-world systems and data. By combining flexibility with guardrails, OpenAI Agent Builder helps teams create scalable, production-ready agentic applications for automation, product assistance, and enterprise operations.

Langchain
LangGraph is an open-source framework for building stateful, multi-agent applications powered by large language models. It enables developers to design, orchestrate, and manage intelligent agents that reason, remember, and collaborate across complex workflows. LangGraph provides a graph-based architecture for creating adaptive AI systems with persistent memory, tool integration, and controllable logic—making it ideal for scalable, production-ready agent development.
