OpenHands AI Coding Agent - Self-Hosted

This is a repackaged open source software product wherein additional charges apply for cloudimg support services.

Why This AMI Over Manual Installation

This image eliminates the multi-step container setup, nginx reverse proxy configuration, storage volume provisioning and credential hardening you would perform yourself when installing OpenHands from source. A private, self hosted agent is running within minutes of launch rather than the hours a manual deployment typically takes.

Overview

OpenHands is an open source AI software engineering agent. You describe a task in a browser, and the agent plans it, then reads and edits files, runs shell commands and browses the web to carry it out, working inside a sandboxed runtime container that it launches on demand. This image delivers OpenHands 1.8 fully installed and configured, so you can start automating engineering work immediately.

Bring Your Own Model Provider Key

No model API key, model credits or hosted inference endpoint ship with this image. OpenHands opens its provider configuration screen on first visit, where you enter your own provider API key, for example Anthropic or OpenAI. This is a required first run step. Your key and all conversation state are stored only on your own instance, and model requests go directly from your instance to the provider you choose, rather than through a third party service. You keep control of which model you use and where your data goes.

Application Stack

OpenHands runs as the official application container under Docker, bound to the instance loopback address, with nginx fronting it on port 80. The agent starts its sandboxed runtime containers through the host Docker socket. State, which means your settings, conversation history and the agent workspace, and the Docker image store both live on their own dedicated, independently resizable storage volumes kept separate from the operating system disk.

Secure By Default

OpenHands has no built in login of its own, and the agent can run commands on the instance, so access is gated by HTTP Basic Authentication at the nginx reverse proxy in front of the whole application. This image generates a fresh password, unique to your instance, on its first boot and writes it to a root only file. No shared or default credentials ship in the image, which removes a common misconfiguration risk from day one. An unauthenticated health endpoint stays open for load balancer probes; everything else requires the password.

Getting Started

1. Launch the AMI on your preferred instance type.

2. Retrieve the unique password from the instance.

3. Open the instance address in your browser and sign in.

4. Enter your model provider API key in the settings screen.

5. Describe your first task and the agent executes it in a sandboxed container.

Use Cases

A private, self hosted coding agent inside your own virtual private cloud, for teams that need code and prompts to stay on infrastructure they control. Automating repetitive engineering work such as refactors, dependency upgrades, test writing and bug reproduction. Evaluating agentic software development on your own repositories, against the model provider you already use.

cloudimg Support

24/7 technical support by email and chat. Help with OpenHands deployment, instance sizing, connecting your own model provider, the sandboxed runtime and Docker configuration, TLS termination and scaling.

All product and company names are trademarks or registered trademarks of their respective holders. Use of them does not imply any affiliation with or endorsement by them.