OpenHands

Overview

OpenHands is an AI agent platform for software development that executes real engineering work, not just suggests code. It runs autonomous agents that plan, write, and apply changes across your codebase, helping teams complete complex tasks end-to-end. Built on an open source foundation, it is designed for enterprise software development at scale.

Key Features

• Model-agnostic: bring your own agent (Claude Code, Codex, Gemini, OpenHands) and any LLM
• Automations triggered from GitHub, Slack, Jira, Linear, schedules, and webhooks
• Self-hosted or cloud deployment with secure sandboxes and full auditability
• Multi-backend support: run agents locally, in Docker, on VMs, or in your VPC
• Agent Control Plane for centralized governance of who, what, where, and spend
• Open source core with MIT license and 89.2K GitHub stars

Use Cases

• Review pull requests and fix CI failures automatically
• Triage issues and open draft PRs with context
• Automate QA across the release cycle
• Investigate incidents and propose patches

Getting Started

• Install via npm: npm install -g @openhands/agent-canvas
• Run agent-canvas to start the local stack
• Optionally use Docker sandbox for isolated execution

Deployment & Requirements

• Node.js 24 or later and uv for local installation
• Docker for sandboxed execution
• Can run on laptop, dedicated machine, or cloud server

Before You Adopt

• License: MIT. Review its terms before using, modifying, or distributing the project.
• Running without a sandbox gives the agent full access to your filesystem.
• Self-hosting ensures your code never leaves your control.