Skip to main content

​
What is HAPI?

HAPI is a local-first platform that lets you run official AI coding agents (Claude Code, Codex, Cursor Agent, Gemini, OpenCode) locally and control them remotely through a web app, PWA, or Telegram Mini App. Never lose context, switch between local and remote seamlessly, and approve AI requests from anywhere.

Quick Start

Get up and running in 2 minutes with HAPI

Installation

Install HAPI on your machine

Architecture

Understand how HAPI works under the hood

CLI Reference

Explore all CLI commands and options

​
Key Features

Seamless Handoff

Work locally, switch to remote when needed, switch back anytime. No context loss, no session restart.

Native First

HAPI wraps your AI agent instead of replacing it. Same terminal, same experience, same muscle memory.

Remote Control

Approve AI requests from your phone or browser with one tap. Step away from your desk without stopping your work.

Multi-Agent Support

Claude Code, Codex, Cursor Agent, Gemini, OpenCode—different models, one unified workflow.

Terminal Anywhere

Run commands from your phone or browser, directly connected to the working machine.

Voice Control

Talk to your AI agent hands-free using the built-in voice assistant.

​
Supported Agents

HAPI supports all major AI coding agents:

Claude Code

Official Claude Code integration

Codex

OpenAI Codex mode

Cursor Agent

Cursor CLI integration

Gemini

Gemini via ACP

OpenCode

OpenCode via ACP

​
Architecture Overview

HAPI consists of three main components that work together:
1

CLI

Wraps your AI agent and connects to the hub via Socket.IO for real-time communication
2

Hub

HTTP API + Socket.IO server that persists sessions in SQLite and broadcasts updates via SSE
3

Web App

React PWA for remote control with live updates and Telegram Mini App integration
All data stays on your machine. The relay uses WireGuard + TLS for end-to-end encryption when remote access is needed.
Start a coding session on your desktop, review and approve changes from your phone during lunch, then resume from your laptop.
Let your AI agent work on large refactors while you’re in meetings. Get notifications when approval is needed.
Share sessions with team members using namespaces. Each developer can run their own agents while maintaining isolation.
Use the voice assistant to give instructions hands-free while working on other tasks.

​
Getting Started

Quickstart Guide

Start using HAPI in under 2 minutes

Core Concepts

Learn how HAPI works

Deployment Options

Choose your deployment method

API Reference

Integrate with HAPI’s API

Build docs developers (and LLMs) love

Get started for freeTalk to us